End of training

.gitattributes

@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
+result_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -4,6 +4,7 @@ license: other
 base_model: mistralai/Ministral-8B-Instruct-2410
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 metrics:
 - accuracy
@@ -17,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Ministral-8B-Instruct-2410-SFT-envbench_weave_2500
 
-This model is a fine-tuned version of [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) on the envbench_weave_2500 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4115
-- Accuracy: 0.8912
-- Num Input Tokens Seen: 87699168
+- Loss: 0.4081
+- Accuracy: 0.8916
+- Num Input Tokens Seen: 95021568
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.8915727021499925,
+    "eval_loss": 0.40814870595932007,
+    "eval_runtime": 9.8343,
+    "eval_samples_per_second": 12.711,
+    "eval_steps_per_second": 0.813,
+    "num_input_tokens_seen": 95021568,
+    "total_flos": 4.266242635254989e+18,
+    "train_loss": 0.7843870968410843,
+    "train_runtime": 5432.7802,
+    "train_samples_per_second": 2.186,
+    "train_steps_per_second": 0.035
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.8915727021499925,
+    "eval_loss": 0.40814870595932007,
+    "eval_runtime": 9.8343,
+    "eval_samples_per_second": 12.711,
+    "eval_steps_per_second": 0.813,
+    "num_input_tokens_seen": 95021568
+}

result_model/README.md

@@ -0,0 +1,66 @@
+---
+library_name: transformers
+license: other
+base_model: mistralai/Ministral-8B-Instruct-2410
+tags:
+- llama-factory
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: Ministral-8B-Instruct-2410-SFT-envbench_weave_2500
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# Ministral-8B-Instruct-2410-SFT-envbench_weave_2500
+
+This model is a fine-tuned version of [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4115
+- Accuracy: 0.8912
+- Num Input Tokens Seen: 87699168
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 5.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.52.4
+- Pytorch 2.6.0a0+df5bbc09d1.nv24.12
+- Datasets 3.6.0
+- Tokenizers 0.21.1

result_model/all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "num_input_tokens_seen": 95021568,
+    "total_flos": 4.266242635254989e+18,
+    "train_loss": 0.7843870968410843,
+    "train_runtime": 5432.7802,
+    "train_samples_per_second": 2.186,
+    "train_steps_per_second": 0.035
+}

result_model/chat_template.jinja

@@ -0,0 +1,87 @@
+{%- if messages[0]["role"] == "system" %}
+    {%- set system_message = messages[0]["content"] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = none %}
+{%- endif %}
+{%- set user_messages = loop_messages | selectattr("role", "equalto", "user") | list %}
+
+{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}
+{%- set ns = namespace() %}
+{%- set ns.index = 0 %}
+{%- for message in loop_messages %}
+    {%- if not (message.role == "tool" or message.role == "tool_results" or (message.tool_calls is defined and message.tool_calls is not none)) %}
+        {%- if (message["role"] == "user") != (ns.index % 2 == 0) %}
+            {{- raise_exception("After the optional system message, conversation roles must alternate user/assistant/user/assistant/...") }}
+        {%- endif %}
+        {%- set ns.index = ns.index + 1 %}
+    {%- endif %}
+{%- endfor %}
+
+{{- bos_token }}
+{%- for message in loop_messages %}
+    {%- if message["role"] == "user" %}
+        {%- if tools is not none and (message == user_messages[-1]) %}
+            {{- "[AVAILABLE_TOOLS][" }}
+            {%- for tool in tools %}
+                {%- set tool = tool.function %}
+                {{- '{"type": "function", "function": {' }}
+                {%- for key, val in tool.items() if key != "return" %}
+                    {%- if val is string %}
+                        {{- '"' + key + '": "' + val + '"' }}
+                    {%- else %}
+                        {{- '"' + key + '": ' + val|tojson }}
+                    {%- endif %}
+                    {%- if not loop.last %}
+                        {{- ", " }}
+                    {%- endif %}
+                {%- endfor %}
+                {{- "}}" }}
+                {%- if not loop.last %}
+                    {{- ", " }}
+                {%- else %}
+                    {{- "]" }}
+                {%- endif %}
+            {%- endfor %}
+            {{- "[/AVAILABLE_TOOLS]" }}
+            {%- endif %}
+        {%- if loop.last and system_message is defined %}
+            {{- "[INST]" + system_message + "\n\n" + message["content"] + "[/INST]" }}
+        {%- else %}
+            {{- "[INST]" + message["content"] + "[/INST]" }}
+        {%- endif %}
+    {%- elif (message.tool_calls is defined and message.tool_calls is not none) %}
+        {{- "[TOOL_CALLS][" }}
+        {%- for tool_call in message.tool_calls %}
+            {%- set out = tool_call.function|tojson %}
+            {{- out[:-1] }}
+            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}
+                {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
+            {%- endif %}
+            {{- ', "id": "' + tool_call.id + '"}' }}
+            {%- if not loop.last %}
+                {{- ", " }}
+            {%- else %}
+                {{- "]" + eos_token }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif message["role"] == "assistant" %}
+        {{- message["content"] + eos_token}}
+    {%- elif message["role"] == "tool_results" or message["role"] == "tool" %}
+        {%- if message.content is defined and message.content.content is defined %}
+            {%- set content = message.content.content %}
+        {%- else %}
+            {%- set content = message.content %}
+        {%- endif %}
+        {{- '[TOOL_RESULTS]{"content": ' + content|string + ", " }}
+        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}
+            {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
+        {%- endif %}
+        {{- '"call_id": "' + message.tool_call_id + '"}[/TOOL_RESULTS]' }}
+    {%- else %}
+        {{- raise_exception("Only user and assistant roles are supported, with the exception of an initial optional system message!") }}
+    {%- endif %}
+{%- endfor %}

result_model/config.json

@@ -0,0 +1,64 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 12288,
+  "layer_types": [
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 100000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.52.4",
+  "use_cache": false,
+  "vocab_size": 131072
+}

result_model/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.52.4"
+}

result_model/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b43100279b4a926733ef716fde7237fd10bfc2320c5e46b0d1cf0792a2804f10
+size 4983007904

result_model/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4488fedf35099dd97b9012af4a10b1111f2f45fc0c9b0027738d48aabf268dd5
+size 4999836776

result_model/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85080b2ed21b76a4eed7b46f93d292e7b7b037e0e3a73cb864bbf5ae5071fbab
+size 4983067960

result_model/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:906fc75a681843383a70521a5f569106c46b2b8bc1896c6e04e37b73dcb01fab
+size 1073741952

result_model/model.safetensors.index.json

@@ -0,0 +1,334 @@
+{
+  "metadata": {
+    "total_size": 16039616512
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

result_model/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

result_model/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7edbeaf20dd7f571b5dd1c54d9ace4f9b6299127cc7ba2afb14a6d51a4a79a4
+size 17078136

result_model/train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "num_input_tokens_seen": 95021568,
+    "total_flos": 4.266242635254989e+18,
+    "train_loss": 0.7843870968410843,
+    "train_runtime": 5432.7802,
+    "train_samples_per_second": 2.186,
+    "train_steps_per_second": 0.035
+}

result_model/trainer_log.jsonl

@@ -0,0 +1,198 @@
+{"current_steps": 1, "total_steps": 190, "loss": 0.5587, "lr": 0.0, "epoch": 0.026845637583892617, "percentage": 0.53, "elapsed_time": "0:00:34", "remaining_time": "1:49:01", "throughput": 14331.04, "total_tokens": 496000}
+{"current_steps": 2, "total_steps": 190, "loss": 0.6532, "lr": 2.631578947368421e-06, "epoch": 0.053691275167785234, "percentage": 1.05, "elapsed_time": "0:01:00", "remaining_time": "1:35:17", "throughput": 16561.55, "total_tokens": 1007328}
+{"current_steps": 3, "total_steps": 190, "loss": 0.4746, "lr": 5.263157894736842e-06, "epoch": 0.08053691275167785, "percentage": 1.58, "elapsed_time": "0:01:26", "remaining_time": "1:30:04", "throughput": 17532.25, "total_tokens": 1520160}
+{"current_steps": 4, "total_steps": 190, "loss": 0.5696, "lr": 7.894736842105263e-06, "epoch": 0.10738255033557047, "percentage": 2.11, "elapsed_time": "0:01:52", "remaining_time": "1:27:16", "throughput": 18037.54, "total_tokens": 2031424}
+{"current_steps": 5, "total_steps": 190, "loss": 0.5954, "lr": 1.0526315789473684e-05, "epoch": 0.1342281879194631, "percentage": 2.63, "elapsed_time": "0:02:18", "remaining_time": "1:25:27", "throughput": 18348.79, "total_tokens": 2542624}
+{"current_steps": 6, "total_steps": 190, "loss": 0.7396, "lr": 1.3157894736842106e-05, "epoch": 0.1610738255033557, "percentage": 3.16, "elapsed_time": "0:02:44", "remaining_time": "1:24:00", "throughput": 18591.36, "total_tokens": 3056000}
+{"current_steps": 7, "total_steps": 190, "loss": 0.7118, "lr": 1.5789473684210526e-05, "epoch": 0.18791946308724833, "percentage": 3.68, "elapsed_time": "0:03:10", "remaining_time": "1:22:55", "throughput": 18778.0, "total_tokens": 3574176}
+{"current_steps": 8, "total_steps": 190, "loss": 0.5509, "lr": 1.8421052631578947e-05, "epoch": 0.21476510067114093, "percentage": 4.21, "elapsed_time": "0:03:37", "remaining_time": "1:22:17", "throughput": 18783.0, "total_tokens": 4076224}
+{"current_steps": 9, "total_steps": 190, "loss": 0.9419, "lr": 2.105263157894737e-05, "epoch": 0.24161073825503357, "percentage": 4.74, "elapsed_time": "0:04:02", "remaining_time": "1:21:26", "throughput": 18861.24, "total_tokens": 4582944}
+{"current_steps": 10, "total_steps": 190, "loss": 0.6846, "lr": 2.368421052631579e-05, "epoch": 0.2684563758389262, "percentage": 5.26, "elapsed_time": "0:04:29", "remaining_time": "1:20:46", "throughput": 18881.76, "total_tokens": 5083968}
+{"current_steps": 11, "total_steps": 190, "loss": 0.5596, "lr": 2.6315789473684212e-05, "epoch": 0.2953020134228188, "percentage": 5.79, "elapsed_time": "0:04:55", "remaining_time": "1:20:09", "throughput": 18975.83, "total_tokens": 5608256}
+{"current_steps": 12, "total_steps": 190, "loss": 0.5313, "lr": 2.8947368421052634e-05, "epoch": 0.3221476510067114, "percentage": 6.32, "elapsed_time": "0:05:21", "remaining_time": "1:19:29", "throughput": 19014.41, "total_tokens": 6113696}
+{"current_steps": 13, "total_steps": 190, "loss": 0.7476, "lr": 3.157894736842105e-05, "epoch": 0.348993288590604, "percentage": 6.84, "elapsed_time": "0:05:47", "remaining_time": "1:18:49", "throughput": 19001.04, "total_tokens": 6599840}
+{"current_steps": 14, "total_steps": 190, "loss": 0.6272, "lr": 3.421052631578947e-05, "epoch": 0.37583892617449666, "percentage": 7.37, "elapsed_time": "0:06:13", "remaining_time": "1:18:17", "throughput": 19043.96, "total_tokens": 7116640}
+{"current_steps": 15, "total_steps": 190, "loss": 0.4918, "lr": 3.6842105263157895e-05, "epoch": 0.40268456375838924, "percentage": 7.89, "elapsed_time": "0:06:39", "remaining_time": "1:17:41", "throughput": 19098.42, "total_tokens": 7631552}
+{"current_steps": 16, "total_steps": 190, "loss": 0.5583, "lr": 3.9473684210526316e-05, "epoch": 0.42953020134228187, "percentage": 8.42, "elapsed_time": "0:07:05", "remaining_time": "1:17:09", "throughput": 19096.11, "total_tokens": 8129664}
+{"current_steps": 17, "total_steps": 190, "loss": 0.6327, "lr": 4.210526315789474e-05, "epoch": 0.4563758389261745, "percentage": 8.95, "elapsed_time": "0:07:31", "remaining_time": "1:16:37", "throughput": 19144.93, "total_tokens": 8648320}
+{"current_steps": 18, "total_steps": 190, "loss": 0.5684, "lr": 4.473684210526316e-05, "epoch": 0.48322147651006714, "percentage": 9.47, "elapsed_time": "0:07:57", "remaining_time": "1:16:06", "throughput": 19181.08, "total_tokens": 9166688}
+{"current_steps": 19, "total_steps": 190, "loss": 0.6136, "lr": 4.736842105263158e-05, "epoch": 0.5100671140939598, "percentage": 10.0, "elapsed_time": "0:08:23", "remaining_time": "1:15:34", "throughput": 19234.07, "total_tokens": 9689760}
+{"current_steps": 20, "total_steps": 190, "loss": 0.6268, "lr": 5e-05, "epoch": 0.5369127516778524, "percentage": 10.53, "elapsed_time": "0:08:50", "remaining_time": "1:15:05", "throughput": 19248.26, "total_tokens": 10202784}
+{"current_steps": 21, "total_steps": 190, "loss": 0.6514, "lr": 4.999578104083307e-05, "epoch": 0.5637583892617449, "percentage": 11.05, "elapsed_time": "0:09:16", "remaining_time": "1:14:34", "throughput": 19252.73, "total_tokens": 10704736}
+{"current_steps": 22, "total_steps": 190, "loss": 0.4698, "lr": 4.998312558730159e-05, "epoch": 0.5906040268456376, "percentage": 11.58, "elapsed_time": "0:09:42", "remaining_time": "1:14:05", "throughput": 19245.25, "total_tokens": 11204608}
+{"current_steps": 23, "total_steps": 190, "loss": 0.6816, "lr": 4.996203791083291e-05, "epoch": 0.6174496644295302, "percentage": 12.11, "elapsed_time": "0:10:08", "remaining_time": "1:13:36", "throughput": 19238.29, "total_tokens": 11702432}
+{"current_steps": 24, "total_steps": 190, "loss": 0.6014, "lr": 4.993252512887069e-05, "epoch": 0.6442953020134228, "percentage": 12.63, "elapsed_time": "0:10:34", "remaining_time": "1:13:09", "throughput": 19238.38, "total_tokens": 12210336}
+{"current_steps": 25, "total_steps": 190, "loss": 2.6525, "lr": 4.9894597202472696e-05, "epoch": 0.6711409395973155, "percentage": 13.16, "elapsed_time": "0:11:00", "remaining_time": "1:12:39", "throughput": 19264.33, "total_tokens": 12725728}
+{"current_steps": 25, "total_steps": 190, "eval_loss": 0.7538173198699951, "epoch": 0.6711409395973155, "percentage": 13.16, "elapsed_time": "0:11:10", "remaining_time": "1:13:46", "throughput": 18975.16, "total_tokens": 12725728}
+{"current_steps": 26, "total_steps": 190, "loss": 3.0007, "lr": 4.984826693294874e-05, "epoch": 0.697986577181208, "percentage": 13.68, "elapsed_time": "0:12:41", "remaining_time": "1:20:05", "throughput": 17381.69, "total_tokens": 13241920}
+{"current_steps": 27, "total_steps": 190, "loss": 1.1118, "lr": 4.979354995754006e-05, "epoch": 0.7248322147651006, "percentage": 14.21, "elapsed_time": "0:13:07", "remaining_time": "1:19:16", "throughput": 17456.83, "total_tokens": 13754432}
+{"current_steps": 28, "total_steps": 190, "loss": 5.0379, "lr": 4.9730464744141445e-05, "epoch": 0.7516778523489933, "percentage": 14.74, "elapsed_time": "0:13:33", "remaining_time": "1:18:28", "throughput": 17532.84, "total_tokens": 14269824}
+{"current_steps": 29, "total_steps": 190, "loss": 1.1003, "lr": 4.965903258506806e-05, "epoch": 0.7785234899328859, "percentage": 15.26, "elapsed_time": "0:13:59", "remaining_time": "1:17:43", "throughput": 17568.32, "total_tokens": 14756000}
+{"current_steps": 30, "total_steps": 190, "loss": 1.1593, "lr": 4.957927758986888e-05, "epoch": 0.8053691275167785, "percentage": 15.79, "elapsed_time": "0:14:26", "remaining_time": "1:16:59", "throughput": 17613.94, "total_tokens": 15256512}
+{"current_steps": 31, "total_steps": 190, "loss": 0.6791, "lr": 4.949122667718935e-05, "epoch": 0.8322147651006712, "percentage": 16.32, "elapsed_time": "0:14:52", "remaining_time": "1:16:15", "throughput": 17686.98, "total_tokens": 15778720}
+{"current_steps": 32, "total_steps": 190, "loss": 0.8403, "lr": 4.9394909565685894e-05, "epoch": 0.8590604026845637, "percentage": 16.84, "elapsed_time": "0:15:17", "remaining_time": "1:15:31", "throughput": 17749.19, "total_tokens": 16291104}
+{"current_steps": 33, "total_steps": 190, "loss": 0.6432, "lr": 4.929035876399535e-05, "epoch": 0.8859060402684564, "percentage": 17.37, "elapsed_time": "0:15:43", "remaining_time": "1:14:49", "throughput": 17814.41, "total_tokens": 16811968}
+{"current_steps": 34, "total_steps": 190, "loss": 0.5564, "lr": 4.917760955976277e-05, "epoch": 0.912751677852349, "percentage": 17.89, "elapsed_time": "0:16:09", "remaining_time": "1:14:10", "throughput": 17864.61, "total_tokens": 17327040}
+{"current_steps": 35, "total_steps": 190, "loss": 1.8105, "lr": 4.905670000773126e-05, "epoch": 0.9395973154362416, "percentage": 18.42, "elapsed_time": "0:16:36", "remaining_time": "1:13:30", "throughput": 17894.73, "total_tokens": 17823168}
+{"current_steps": 36, "total_steps": 190, "loss": 0.9537, "lr": 4.892767091689786e-05, "epoch": 0.9664429530201343, "percentage": 18.95, "elapsed_time": "0:17:02", "remaining_time": "1:12:53", "throughput": 17943.99, "total_tokens": 18343840}
+{"current_steps": 37, "total_steps": 190, "loss": 0.7753, "lr": 4.87905658367398e-05, "epoch": 0.9932885906040269, "percentage": 19.47, "elapsed_time": "0:17:28", "remaining_time": "1:12:14", "throughput": 17995.74, "total_tokens": 18864096}
+{"current_steps": 38, "total_steps": 190, "loss": 0.3534, "lr": 4.864543104251587e-05, "epoch": 1.0, "percentage": 20.0, "elapsed_time": "0:17:33", "remaining_time": "1:10:12", "throughput": 18019.88, "total_tokens": 18979328}
+{"current_steps": 39, "total_steps": 190, "loss": 1.6501, "lr": 4.849231551964771e-05, "epoch": 1.0268456375838926, "percentage": 20.53, "elapsed_time": "0:17:59", "remaining_time": "1:09:39", "throughput": 18041.79, "total_tokens": 19477504}
+{"current_steps": 40, "total_steps": 190, "loss": 2.3538, "lr": 4.833127094718643e-05, "epoch": 1.0536912751677852, "percentage": 21.05, "elapsed_time": "0:18:25", "remaining_time": "1:09:05", "throughput": 18079.89, "total_tokens": 19985312}
+{"current_steps": 41, "total_steps": 190, "loss": 0.5087, "lr": 4.8162351680370044e-05, "epoch": 1.0805369127516777, "percentage": 21.58, "elapsed_time": "0:18:51", "remaining_time": "1:08:32", "throughput": 18096.98, "total_tokens": 20476800}
+{"current_steps": 42, "total_steps": 190, "loss": 0.5662, "lr": 4.79856147322777e-05, "epoch": 1.1073825503355705, "percentage": 22.11, "elapsed_time": "0:19:17", "remaining_time": "1:07:59", "throughput": 18124.39, "total_tokens": 20980288}
+{"current_steps": 43, "total_steps": 190, "loss": 0.4659, "lr": 4.7801119754586766e-05, "epoch": 1.1342281879194631, "percentage": 22.63, "elapsed_time": "0:19:43", "remaining_time": "1:07:26", "throughput": 18159.28, "total_tokens": 21492736}
+{"current_steps": 44, "total_steps": 190, "loss": 1.9776, "lr": 4.760892901743944e-05, "epoch": 1.1610738255033557, "percentage": 23.16, "elapsed_time": "0:20:09", "remaining_time": "1:06:53", "throughput": 18185.9, "total_tokens": 21997856}
+{"current_steps": 45, "total_steps": 190, "loss": 3.0839, "lr": 4.7409107388425504e-05, "epoch": 1.1879194630872483, "percentage": 23.68, "elapsed_time": "0:20:35", "remaining_time": "1:06:22", "throughput": 18213.9, "total_tokens": 22509088}
+{"current_steps": 46, "total_steps": 190, "loss": 0.6674, "lr": 4.7201722310688445e-05, "epoch": 1.2147651006711409, "percentage": 24.21, "elapsed_time": "0:21:02", "remaining_time": "1:05:52", "throughput": 18237.35, "total_tokens": 23024384}
+{"current_steps": 47, "total_steps": 190, "loss": 1.0302, "lr": 4.698684378016222e-05, "epoch": 1.2416107382550337, "percentage": 24.74, "elapsed_time": "0:21:28", "remaining_time": "1:05:19", "throughput": 18271.8, "total_tokens": 23536384}
+{"current_steps": 48, "total_steps": 190, "loss": 1.3633, "lr": 4.676454432194656e-05, "epoch": 1.2684563758389262, "percentage": 25.26, "elapsed_time": "0:21:54", "remaining_time": "1:04:48", "throughput": 18293.44, "total_tokens": 24046432}
+{"current_steps": 49, "total_steps": 190, "loss": 0.5103, "lr": 4.6534898965828405e-05, "epoch": 1.2953020134228188, "percentage": 25.79, "elapsed_time": "0:22:20", "remaining_time": "1:04:17", "throughput": 18311.25, "total_tokens": 24549600}
+{"current_steps": 50, "total_steps": 190, "loss": 0.6796, "lr": 4.629798522095818e-05, "epoch": 1.3221476510067114, "percentage": 26.32, "elapsed_time": "0:22:46", "remaining_time": "1:03:46", "throughput": 18321.09, "total_tokens": 25037312}
+{"current_steps": 50, "total_steps": 190, "eval_loss": 0.5981780886650085, "epoch": 1.3221476510067114, "percentage": 26.32, "elapsed_time": "0:22:56", "remaining_time": "1:04:14", "throughput": 18186.97, "total_tokens": 25037312}
+{"current_steps": 51, "total_steps": 190, "loss": 0.7632, "lr": 4.6053883049689145e-05, "epoch": 1.348993288590604, "percentage": 26.84, "elapsed_time": "0:24:21", "remaining_time": "1:06:24", "throughput": 17475.43, "total_tokens": 25548352}
+{"current_steps": 52, "total_steps": 190, "loss": 0.5686, "lr": 4.580267484058876e-05, "epoch": 1.3758389261744965, "percentage": 27.37, "elapsed_time": "0:24:48", "remaining_time": "1:05:49", "throughput": 17512.79, "total_tokens": 26063136}
+{"current_steps": 53, "total_steps": 190, "loss": 0.5809, "lr": 4.554444538063113e-05, "epoch": 1.4026845637583891, "percentage": 27.89, "elapsed_time": "0:25:14", "remaining_time": "1:05:15", "throughput": 17545.83, "total_tokens": 26574336}
+{"current_steps": 54, "total_steps": 190, "loss": 0.4817, "lr": 4.5279281826580056e-05, "epoch": 1.429530201342282, "percentage": 28.42, "elapsed_time": "0:25:40", "remaining_time": "1:04:40", "throughput": 17572.61, "total_tokens": 27078496}
+{"current_steps": 55, "total_steps": 190, "loss": 0.4111, "lr": 4.5007273675572104e-05, "epoch": 1.4563758389261745, "percentage": 28.95, "elapsed_time": "0:26:07", "remaining_time": "1:04:06", "throughput": 17609.9, "total_tokens": 27599232}
+{"current_steps": 56, "total_steps": 190, "loss": 0.5102, "lr": 4.4728512734909844e-05, "epoch": 1.483221476510067, "percentage": 29.47, "elapsed_time": "0:26:33", "remaining_time": "1:03:33", "throughput": 17633.69, "total_tokens": 28099968}
+{"current_steps": 57, "total_steps": 190, "loss": 0.8591, "lr": 4.444309309107535e-05, "epoch": 1.5100671140939599, "percentage": 30.0, "elapsed_time": "0:26:59", "remaining_time": "1:02:59", "throughput": 17658.84, "total_tokens": 28604864}
+{"current_steps": 58, "total_steps": 190, "loss": 0.6124, "lr": 4.415111107797445e-05, "epoch": 1.5369127516778525, "percentage": 30.53, "elapsed_time": "0:27:26", "remaining_time": "1:02:26", "throughput": 17686.77, "total_tokens": 29112480}
+{"current_steps": 59, "total_steps": 190, "loss": 0.8082, "lr": 4.385266524442241e-05, "epoch": 1.563758389261745, "percentage": 31.05, "elapsed_time": "0:27:51", "remaining_time": "1:01:51", "throughput": 17712.05, "total_tokens": 29610208}
+{"current_steps": 60, "total_steps": 190, "loss": 0.5294, "lr": 4.3547856320882044e-05, "epoch": 1.5906040268456376, "percentage": 31.58, "elapsed_time": "0:28:17", "remaining_time": "1:01:18", "throughput": 17744.08, "total_tokens": 30121920}
+{"current_steps": 61, "total_steps": 190, "loss": 0.5192, "lr": 4.3236787185465525e-05, "epoch": 1.6174496644295302, "percentage": 32.11, "elapsed_time": "0:28:43", "remaining_time": "1:00:44", "throughput": 17773.58, "total_tokens": 30632928}
+{"current_steps": 62, "total_steps": 190, "loss": 0.554, "lr": 4.2919562829211283e-05, "epoch": 1.6442953020134228, "percentage": 32.63, "elapsed_time": "0:29:09", "remaining_time": "1:00:12", "throughput": 17797.41, "total_tokens": 31139872}
+{"current_steps": 63, "total_steps": 190, "loss": 0.4231, "lr": 4.259629032064779e-05, "epoch": 1.6711409395973154, "percentage": 33.16, "elapsed_time": "0:29:35", "remaining_time": "0:59:39", "throughput": 17822.95, "total_tokens": 31645376}
+{"current_steps": 64, "total_steps": 190, "loss": 1.475, "lr": 4.226707876965611e-05, "epoch": 1.697986577181208, "percentage": 33.68, "elapsed_time": "0:30:02", "remaining_time": "0:59:07", "throughput": 17850.53, "total_tokens": 32167168}
+{"current_steps": 65, "total_steps": 190, "loss": 1.6005, "lr": 4.193203929064353e-05, "epoch": 1.7248322147651005, "percentage": 34.21, "elapsed_time": "0:30:28", "remaining_time": "0:58:35", "throughput": 17880.98, "total_tokens": 32691456}
+{"current_steps": 66, "total_steps": 190, "loss": 1.1018, "lr": 4.159128496504053e-05, "epoch": 1.7516778523489933, "percentage": 34.74, "elapsed_time": "0:30:54", "remaining_time": "0:58:04", "throughput": 17894.87, "total_tokens": 33188672}
+{"current_steps": 67, "total_steps": 190, "loss": 1.2071, "lr": 4.1244930803134e-05, "epoch": 1.778523489932886, "percentage": 35.26, "elapsed_time": "0:31:20", "remaining_time": "0:57:33", "throughput": 17911.2, "total_tokens": 33690880}
+{"current_steps": 68, "total_steps": 190, "loss": 0.6214, "lr": 4.089309370524921e-05, "epoch": 1.8053691275167785, "percentage": 35.79, "elapsed_time": "0:31:47", "remaining_time": "0:57:01", "throughput": 17931.64, "total_tokens": 34196736}
+{"current_steps": 69, "total_steps": 190, "loss": 0.5573, "lr": 4.053589242229412e-05, "epoch": 1.8322147651006713, "percentage": 36.32, "elapsed_time": "0:32:13", "remaining_time": "0:56:30", "throughput": 17960.71, "total_tokens": 34720768}
+{"current_steps": 70, "total_steps": 190, "loss": 0.436, "lr": 4.0173447515678916e-05, "epoch": 1.8590604026845639, "percentage": 36.84, "elapsed_time": "0:32:39", "remaining_time": "0:55:58", "throughput": 17983.48, "total_tokens": 35230112}
+{"current_steps": 71, "total_steps": 190, "loss": 0.4724, "lr": 3.9805881316624506e-05, "epoch": 1.8859060402684564, "percentage": 37.37, "elapsed_time": "0:33:05", "remaining_time": "0:55:27", "throughput": 18009.4, "total_tokens": 35752064}
+{"current_steps": 72, "total_steps": 190, "loss": 0.4414, "lr": 3.9433317884873664e-05, "epoch": 1.912751677852349, "percentage": 37.89, "elapsed_time": "0:33:31", "remaining_time": "0:54:57", "throughput": 18030.58, "total_tokens": 36274496}
+{"current_steps": 73, "total_steps": 190, "loss": 0.3534, "lr": 3.905588296681886e-05, "epoch": 1.9395973154362416, "percentage": 38.42, "elapsed_time": "0:33:58", "remaining_time": "0:54:26", "throughput": 18043.29, "total_tokens": 36777152}
+{"current_steps": 74, "total_steps": 190, "loss": 1.1611, "lr": 3.867370395306068e-05, "epoch": 1.9664429530201342, "percentage": 38.95, "elapsed_time": "0:34:24", "remaining_time": "0:53:56", "throughput": 18066.46, "total_tokens": 37298112}
+{"current_steps": 75, "total_steps": 190, "loss": 0.6056, "lr": 3.82869098354114e-05, "epoch": 1.9932885906040267, "percentage": 39.47, "elapsed_time": "0:34:51", "remaining_time": "0:53:26", "throughput": 18082.19, "total_tokens": 37817120}
+{"current_steps": 75, "total_steps": 190, "eval_loss": 0.5577213764190674, "epoch": 1.9932885906040267, "percentage": 39.47, "elapsed_time": "0:35:01", "remaining_time": "0:53:42", "throughput": 17995.52, "total_tokens": 37817120}
+{"current_steps": 76, "total_steps": 190, "loss": 0.4723, "lr": 3.7895631163358105e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "0:36:06", "remaining_time": "0:54:09", "throughput": 17518.61, "total_tokens": 37948192}
+{"current_steps": 77, "total_steps": 190, "loss": 0.4424, "lr": 3.7500000000000003e-05, "epoch": 2.0268456375838926, "percentage": 40.53, "elapsed_time": "0:36:32", "remaining_time": "0:53:37", "throughput": 17545.81, "total_tokens": 38467968}
+{"current_steps": 78, "total_steps": 190, "loss": 1.4039, "lr": 3.7100149877474974e-05, "epoch": 2.053691275167785, "percentage": 41.05, "elapsed_time": "0:36:58", "remaining_time": "0:53:06", "throughput": 17570.6, "total_tokens": 38986496}
+{"current_steps": 79, "total_steps": 190, "loss": 0.3641, "lr": 3.66962157518902e-05, "epoch": 2.0805369127516777, "percentage": 41.58, "elapsed_time": "0:37:25", "remaining_time": "0:52:35", "throughput": 17593.63, "total_tokens": 39507008}
+{"current_steps": 80, "total_steps": 190, "loss": 0.3558, "lr": 3.628833395777224e-05, "epoch": 2.1073825503355703, "percentage": 42.11, "elapsed_time": "0:37:51", "remaining_time": "0:52:03", "throughput": 17613.47, "total_tokens": 40012992}
+{"current_steps": 81, "total_steps": 190, "loss": 0.4281, "lr": 3.587664216205183e-05, "epoch": 2.134228187919463, "percentage": 42.63, "elapsed_time": "0:38:18", "remaining_time": "0:51:32", "throughput": 17639.74, "total_tokens": 40536608}
+{"current_steps": 82, "total_steps": 190, "loss": 0.3695, "lr": 3.546127931759903e-05, "epoch": 2.1610738255033555, "percentage": 43.16, "elapsed_time": "0:38:43", "remaining_time": "0:51:00", "throughput": 17664.76, "total_tokens": 41052192}
+{"current_steps": 83, "total_steps": 190, "loss": 0.967, "lr": 3.504238561632424e-05, "epoch": 2.1879194630872485, "percentage": 43.68, "elapsed_time": "0:39:10", "remaining_time": "0:50:29", "throughput": 17683.82, "total_tokens": 41559584}
+{"current_steps": 84, "total_steps": 190, "loss": 0.4472, "lr": 3.4620102441861143e-05, "epoch": 2.214765100671141, "percentage": 44.21, "elapsed_time": "0:39:36", "remaining_time": "0:49:58", "throughput": 17700.74, "total_tokens": 42060128}
+{"current_steps": 85, "total_steps": 190, "loss": 1.2018, "lr": 3.4194572321847336e-05, "epoch": 2.2416107382550337, "percentage": 44.74, "elapsed_time": "0:40:02", "remaining_time": "0:49:27", "throughput": 17719.72, "total_tokens": 42570080}
+{"current_steps": 86, "total_steps": 190, "loss": 1.1866, "lr": 3.376593887981887e-05, "epoch": 2.2684563758389262, "percentage": 45.26, "elapsed_time": "0:40:28", "remaining_time": "0:48:56", "throughput": 17733.69, "total_tokens": 43068832}
+{"current_steps": 87, "total_steps": 190, "loss": 1.9914, "lr": 3.333434678673489e-05, "epoch": 2.295302013422819, "percentage": 45.79, "elapsed_time": "0:40:54", "remaining_time": "0:48:25", "throughput": 17756.43, "total_tokens": 43582016}
+{"current_steps": 88, "total_steps": 190, "loss": 0.6787, "lr": 3.289994171214882e-05, "epoch": 2.3221476510067114, "percentage": 46.32, "elapsed_time": "0:41:20", "remaining_time": "0:47:55", "throughput": 17773.06, "total_tokens": 44089344}
+{"current_steps": 89, "total_steps": 190, "loss": 0.4743, "lr": 3.246287027504237e-05, "epoch": 2.348993288590604, "percentage": 46.84, "elapsed_time": "0:41:46", "remaining_time": "0:47:24", "throughput": 17790.33, "total_tokens": 44589408}
+{"current_steps": 90, "total_steps": 190, "loss": 0.4042, "lr": 3.202327999433924e-05, "epoch": 2.3758389261744965, "percentage": 47.37, "elapsed_time": "0:42:12", "remaining_time": "0:46:53", "throughput": 17812.0, "total_tokens": 45105888}
+{"current_steps": 91, "total_steps": 190, "loss": 0.4187, "lr": 3.158131923911498e-05, "epoch": 2.402684563758389, "percentage": 47.89, "elapsed_time": "0:42:38", "remaining_time": "0:46:23", "throughput": 17831.49, "total_tokens": 45623968}
+{"current_steps": 92, "total_steps": 190, "loss": 3.5044, "lr": 3.1137137178519985e-05, "epoch": 2.4295302013422817, "percentage": 48.42, "elapsed_time": "0:43:04", "remaining_time": "0:45:53", "throughput": 17849.29, "total_tokens": 46136544}
+{"current_steps": 93, "total_steps": 190, "loss": 0.3795, "lr": 3.069088373143234e-05, "epoch": 2.4563758389261743, "percentage": 48.95, "elapsed_time": "0:43:30", "remaining_time": "0:45:23", "throughput": 17867.58, "total_tokens": 46651040}
+{"current_steps": 94, "total_steps": 190, "loss": 2.3593, "lr": 3.0242709515857758e-05, "epoch": 2.4832214765100673, "percentage": 49.47, "elapsed_time": "0:43:57", "remaining_time": "0:44:53", "throughput": 17883.55, "total_tokens": 47163616}
+{"current_steps": 95, "total_steps": 190, "loss": 1.9348, "lr": 2.9792765798093465e-05, "epoch": 2.51006711409396, "percentage": 50.0, "elapsed_time": "0:44:23", "remaining_time": "0:44:23", "throughput": 17902.1, "total_tokens": 47674976}
+{"current_steps": 96, "total_steps": 190, "loss": 5.973, "lr": 2.9341204441673266e-05, "epoch": 2.5369127516778525, "percentage": 50.53, "elapsed_time": "0:44:49", "remaining_time": "0:43:53", "throughput": 17914.85, "total_tokens": 48185504}
+{"current_steps": 97, "total_steps": 190, "loss": 0.5242, "lr": 2.8888177856111083e-05, "epoch": 2.563758389261745, "percentage": 51.05, "elapsed_time": "0:45:15", "remaining_time": "0:43:23", "throughput": 17928.93, "total_tokens": 48689920}
+{"current_steps": 98, "total_steps": 190, "loss": 0.6778, "lr": 2.8433838945460205e-05, "epoch": 2.5906040268456376, "percentage": 51.58, "elapsed_time": "0:45:41", "remaining_time": "0:42:53", "throughput": 17948.81, "total_tokens": 49207680}
+{"current_steps": 99, "total_steps": 190, "loss": 0.4994, "lr": 2.797834105670559e-05, "epoch": 2.61744966442953, "percentage": 52.11, "elapsed_time": "0:46:07", "remaining_time": "0:42:24", "throughput": 17957.54, "total_tokens": 49701344}
+{"current_steps": 100, "total_steps": 190, "loss": 0.4019, "lr": 2.752183792800671e-05, "epoch": 2.6442953020134228, "percentage": 52.63, "elapsed_time": "0:46:33", "remaining_time": "0:41:54", "throughput": 17974.14, "total_tokens": 50216704}
+{"current_steps": 100, "total_steps": 190, "eval_loss": 0.5278336405754089, "epoch": 2.6442953020134228, "percentage": 52.63, "elapsed_time": "0:46:43", "remaining_time": "0:42:03", "throughput": 17909.57, "total_tokens": 50216704}
+{"current_steps": 101, "total_steps": 190, "loss": 1.1764, "lr": 2.7064483636808313e-05, "epoch": 2.6711409395973154, "percentage": 53.16, "elapsed_time": "0:48:09", "remaining_time": "0:42:26", "throughput": 17553.63, "total_tokens": 50727392}
+{"current_steps": 102, "total_steps": 190, "loss": 0.8064, "lr": 2.6606432547836757e-05, "epoch": 2.697986577181208, "percentage": 53.68, "elapsed_time": "0:48:36", "remaining_time": "0:41:56", "throughput": 17571.22, "total_tokens": 51242496}
+{"current_steps": 103, "total_steps": 190, "loss": 6.5702, "lr": 2.6147839260999295e-05, "epoch": 2.7248322147651005, "percentage": 54.21, "elapsed_time": "0:49:02", "remaining_time": "0:41:25", "throughput": 17580.85, "total_tokens": 51725728}
+{"current_steps": 104, "total_steps": 190, "loss": 0.3993, "lr": 2.5688858559204053e-05, "epoch": 2.751677852348993, "percentage": 54.74, "elapsed_time": "0:49:28", "remaining_time": "0:40:55", "throughput": 17597.18, "total_tokens": 52244192}
+{"current_steps": 105, "total_steps": 190, "loss": 0.4268, "lr": 2.5229645356118163e-05, "epoch": 2.778523489932886, "percentage": 55.26, "elapsed_time": "0:49:55", "remaining_time": "0:40:24", "throughput": 17613.59, "total_tokens": 52757856}
+{"current_steps": 106, "total_steps": 190, "loss": 0.358, "lr": 2.4770354643881843e-05, "epoch": 2.8053691275167782, "percentage": 55.79, "elapsed_time": "0:50:21", "remaining_time": "0:39:54", "throughput": 17629.58, "total_tokens": 53267776}
+{"current_steps": 107, "total_steps": 190, "loss": 1.1325, "lr": 2.4311141440795953e-05, "epoch": 2.8322147651006713, "percentage": 56.32, "elapsed_time": "0:50:47", "remaining_time": "0:39:23", "throughput": 17645.15, "total_tokens": 53774464}
+{"current_steps": 108, "total_steps": 190, "loss": 3.3749, "lr": 2.3852160739000707e-05, "epoch": 2.859060402684564, "percentage": 56.84, "elapsed_time": "0:51:13", "remaining_time": "0:38:53", "throughput": 17663.39, "total_tokens": 54287936}
+{"current_steps": 109, "total_steps": 190, "loss": 0.3589, "lr": 2.3393567452163252e-05, "epoch": 2.8859060402684564, "percentage": 57.37, "elapsed_time": "0:51:39", "remaining_time": "0:38:23", "throughput": 17672.42, "total_tokens": 54776736}
+{"current_steps": 110, "total_steps": 190, "loss": 0.4193, "lr": 2.2935516363191693e-05, "epoch": 2.912751677852349, "percentage": 57.89, "elapsed_time": "0:52:05", "remaining_time": "0:37:53", "throughput": 17683.15, "total_tokens": 55271424}
+{"current_steps": 111, "total_steps": 190, "loss": 0.3847, "lr": 2.2478162071993298e-05, "epoch": 2.9395973154362416, "percentage": 58.42, "elapsed_time": "0:52:31", "remaining_time": "0:37:22", "throughput": 17696.58, "total_tokens": 55769824}
+{"current_steps": 112, "total_steps": 190, "loss": 0.3429, "lr": 2.202165894329441e-05, "epoch": 2.966442953020134, "percentage": 58.95, "elapsed_time": "0:52:57", "remaining_time": "0:36:52", "throughput": 17715.06, "total_tokens": 56289728}
+{"current_steps": 113, "total_steps": 190, "loss": 0.4288, "lr": 2.1566161054539798e-05, "epoch": 2.9932885906040267, "percentage": 59.47, "elapsed_time": "0:53:23", "remaining_time": "0:36:23", "throughput": 17731.51, "total_tokens": 56806016}
+{"current_steps": 114, "total_steps": 190, "loss": 0.389, "lr": 2.111182214388893e-05, "epoch": 3.0, "percentage": 60.0, "elapsed_time": "0:53:28", "remaining_time": "0:35:39", "throughput": 17743.92, "total_tokens": 56937088}
+{"current_steps": 115, "total_steps": 190, "loss": 3.2378, "lr": 2.0658795558326743e-05, "epoch": 3.0268456375838926, "percentage": 60.53, "elapsed_time": "0:53:55", "remaining_time": "0:35:10", "throughput": 17752.98, "total_tokens": 57437376}
+{"current_steps": 116, "total_steps": 190, "loss": 0.7378, "lr": 2.0207234201906547e-05, "epoch": 3.053691275167785, "percentage": 61.05, "elapsed_time": "0:54:21", "remaining_time": "0:34:40", "throughput": 17760.7, "total_tokens": 57926240}
+{"current_steps": 117, "total_steps": 190, "loss": 0.3071, "lr": 1.9757290484142244e-05, "epoch": 3.0805369127516777, "percentage": 61.58, "elapsed_time": "0:54:47", "remaining_time": "0:34:11", "throughput": 17776.9, "total_tokens": 58445376}
+{"current_steps": 118, "total_steps": 190, "loss": 0.5286, "lr": 1.9309116268567674e-05, "epoch": 3.1073825503355703, "percentage": 62.11, "elapsed_time": "0:55:13", "remaining_time": "0:33:41", "throughput": 17788.56, "total_tokens": 58940512}
+{"current_steps": 119, "total_steps": 190, "loss": 1.2341, "lr": 1.8862862821480025e-05, "epoch": 3.134228187919463, "percentage": 62.63, "elapsed_time": "0:55:39", "remaining_time": "0:33:12", "throughput": 17801.65, "total_tokens": 59447232}
+{"current_steps": 120, "total_steps": 190, "loss": 0.2624, "lr": 1.8418680760885027e-05, "epoch": 3.1610738255033555, "percentage": 63.16, "elapsed_time": "0:56:05", "remaining_time": "0:32:43", "throughput": 17816.24, "total_tokens": 59962592}
+{"current_steps": 121, "total_steps": 190, "loss": 0.3812, "lr": 1.797672000566077e-05, "epoch": 3.1879194630872485, "percentage": 63.68, "elapsed_time": "0:56:31", "remaining_time": "0:32:13", "throughput": 17832.25, "total_tokens": 60473632}
+{"current_steps": 122, "total_steps": 190, "loss": 0.3298, "lr": 1.7537129724957642e-05, "epoch": 3.214765100671141, "percentage": 64.21, "elapsed_time": "0:56:57", "remaining_time": "0:31:44", "throughput": 17846.21, "total_tokens": 60984768}
+{"current_steps": 123, "total_steps": 190, "loss": 0.316, "lr": 1.710005828785119e-05, "epoch": 3.2416107382550337, "percentage": 64.74, "elapsed_time": "0:57:23", "remaining_time": "0:31:15", "throughput": 17861.23, "total_tokens": 61507680}
+{"current_steps": 124, "total_steps": 190, "loss": 0.8116, "lr": 1.666565321326512e-05, "epoch": 3.2684563758389262, "percentage": 65.26, "elapsed_time": "0:57:49", "remaining_time": "0:30:46", "throughput": 17875.92, "total_tokens": 62023520}
+{"current_steps": 125, "total_steps": 190, "loss": 0.3133, "lr": 1.6234061120181142e-05, "epoch": 3.295302013422819, "percentage": 65.79, "elapsed_time": "0:58:15", "remaining_time": "0:30:17", "throughput": 17888.79, "total_tokens": 62538624}
+{"current_steps": 125, "total_steps": 190, "eval_loss": 0.4720778465270996, "epoch": 3.295302013422819, "percentage": 65.79, "elapsed_time": "0:58:26", "remaining_time": "0:30:23", "throughput": 17837.39, "total_tokens": 62538624}
+{"current_steps": 126, "total_steps": 190, "loss": 0.8765, "lr": 1.5805427678152677e-05, "epoch": 3.3221476510067114, "percentage": 66.32, "elapsed_time": "0:59:52", "remaining_time": "0:30:24", "throughput": 17550.38, "total_tokens": 63047392}
+{"current_steps": 127, "total_steps": 190, "loss": 0.2751, "lr": 1.5379897558138862e-05, "epoch": 3.348993288590604, "percentage": 66.84, "elapsed_time": "1:00:19", "remaining_time": "0:29:55", "throughput": 17558.39, "total_tokens": 63547936}
+{"current_steps": 128, "total_steps": 190, "loss": 1.2805, "lr": 1.495761438367577e-05, "epoch": 3.3758389261744965, "percentage": 67.37, "elapsed_time": "1:00:45", "remaining_time": "0:29:25", "throughput": 17569.76, "total_tokens": 64051232}
+{"current_steps": 129, "total_steps": 190, "loss": 0.2843, "lr": 1.4538720682400969e-05, "epoch": 3.402684563758389, "percentage": 67.89, "elapsed_time": "1:01:12", "remaining_time": "0:28:56", "throughput": 17576.44, "total_tokens": 64546048}
+{"current_steps": 130, "total_steps": 190, "loss": 0.3334, "lr": 1.4123357837948175e-05, "epoch": 3.4295302013422817, "percentage": 68.42, "elapsed_time": "1:01:38", "remaining_time": "0:28:27", "throughput": 17587.19, "total_tokens": 65047392}
+{"current_steps": 131, "total_steps": 190, "loss": 1.2229, "lr": 1.3711666042227772e-05, "epoch": 3.4563758389261743, "percentage": 68.95, "elapsed_time": "1:02:04", "remaining_time": "0:27:57", "throughput": 17601.89, "total_tokens": 65565600}
+{"current_steps": 132, "total_steps": 190, "loss": 0.4387, "lr": 1.330378424810981e-05, "epoch": 3.4832214765100673, "percentage": 69.47, "elapsed_time": "1:02:31", "remaining_time": "0:27:28", "throughput": 17614.44, "total_tokens": 66075392}
+{"current_steps": 133, "total_steps": 190, "loss": 0.2675, "lr": 1.2899850122525037e-05, "epoch": 3.51006711409396, "percentage": 70.0, "elapsed_time": "1:02:56", "remaining_time": "0:26:58", "throughput": 17631.08, "total_tokens": 66591552}
+{"current_steps": 134, "total_steps": 190, "loss": 0.2816, "lr": 1.2500000000000006e-05, "epoch": 3.5369127516778525, "percentage": 70.53, "elapsed_time": "1:03:22", "remaining_time": "0:26:29", "throughput": 17645.96, "total_tokens": 67106688}
+{"current_steps": 135, "total_steps": 190, "loss": 0.2957, "lr": 1.2104368836641908e-05, "epoch": 3.563758389261745, "percentage": 71.05, "elapsed_time": "1:03:48", "remaining_time": "0:25:59", "throughput": 17659.75, "total_tokens": 67618688}
+{"current_steps": 136, "total_steps": 190, "loss": 2.8051, "lr": 1.1713090164588607e-05, "epoch": 3.5906040268456376, "percentage": 71.58, "elapsed_time": "1:04:15", "remaining_time": "0:25:30", "throughput": 17671.63, "total_tokens": 68129952}
+{"current_steps": 137, "total_steps": 190, "loss": 0.2361, "lr": 1.1326296046939333e-05, "epoch": 3.61744966442953, "percentage": 72.11, "elapsed_time": "1:04:41", "remaining_time": "0:25:01", "throughput": 17684.97, "total_tokens": 68642048}
+{"current_steps": 138, "total_steps": 190, "loss": 0.4118, "lr": 1.0944117033181151e-05, "epoch": 3.6442953020134228, "percentage": 72.63, "elapsed_time": "1:05:07", "remaining_time": "0:24:32", "throughput": 17700.03, "total_tokens": 69161312}
+{"current_steps": 139, "total_steps": 190, "loss": 0.3521, "lr": 1.0566682115126344e-05, "epoch": 3.6711409395973154, "percentage": 73.16, "elapsed_time": "1:05:33", "remaining_time": "0:24:03", "throughput": 17710.48, "total_tokens": 69669024}
+{"current_steps": 140, "total_steps": 190, "loss": 0.3169, "lr": 1.0194118683375503e-05, "epoch": 3.697986577181208, "percentage": 73.68, "elapsed_time": "1:05:59", "remaining_time": "0:23:34", "throughput": 17722.53, "total_tokens": 70169152}
+{"current_steps": 141, "total_steps": 190, "loss": 0.2943, "lr": 9.826552484321087e-06, "epoch": 3.7248322147651005, "percentage": 74.21, "elapsed_time": "1:06:25", "remaining_time": "0:23:05", "throughput": 17736.0, "total_tokens": 70691424}
+{"current_steps": 142, "total_steps": 190, "loss": 0.4079, "lr": 9.464107577705886e-06, "epoch": 3.751677852348993, "percentage": 74.74, "elapsed_time": "1:06:51", "remaining_time": "0:22:36", "throughput": 17746.64, "total_tokens": 71198976}
+{"current_steps": 143, "total_steps": 190, "loss": 0.2099, "lr": 9.106906294750805e-06, "epoch": 3.778523489932886, "percentage": 75.26, "elapsed_time": "1:07:18", "remaining_time": "0:22:07", "throughput": 17758.63, "total_tokens": 71713312}
+{"current_steps": 144, "total_steps": 190, "loss": 0.3496, "lr": 8.755069196866014e-06, "epoch": 3.8053691275167782, "percentage": 75.79, "elapsed_time": "1:07:44", "remaining_time": "0:21:38", "throughput": 17771.5, "total_tokens": 72232064}
+{"current_steps": 145, "total_steps": 190, "loss": 0.2696, "lr": 8.40871503495947e-06, "epoch": 3.8322147651006713, "percentage": 76.32, "elapsed_time": "1:08:10", "remaining_time": "0:21:09", "throughput": 17783.18, "total_tokens": 72743648}
+{"current_steps": 146, "total_steps": 190, "loss": 0.2302, "lr": 8.067960709356478e-06, "epoch": 3.859060402684564, "percentage": 76.84, "elapsed_time": "1:08:37", "remaining_time": "0:20:40", "throughput": 17795.87, "total_tokens": 73266496}
+{"current_steps": 147, "total_steps": 190, "loss": 0.9783, "lr": 7.732921230343892e-06, "epoch": 3.8859060402684564, "percentage": 77.37, "elapsed_time": "1:09:03", "remaining_time": "0:20:12", "throughput": 17805.81, "total_tokens": 73779232}
+{"current_steps": 148, "total_steps": 190, "loss": 0.3549, "lr": 7.403709679352217e-06, "epoch": 3.912751677852349, "percentage": 77.89, "elapsed_time": "1:09:29", "remaining_time": "0:19:43", "throughput": 17817.31, "total_tokens": 74283712}
+{"current_steps": 149, "total_steps": 190, "loss": 0.303, "lr": 7.080437170788723e-06, "epoch": 3.9395973154362416, "percentage": 78.42, "elapsed_time": "1:09:55", "remaining_time": "0:19:14", "throughput": 17827.21, "total_tokens": 74794496}
+{"current_steps": 150, "total_steps": 190, "loss": 0.3139, "lr": 6.763212814534484e-06, "epoch": 3.966442953020134, "percentage": 78.95, "elapsed_time": "1:10:21", "remaining_time": "0:18:45", "throughput": 17838.26, "total_tokens": 75296576}
+{"current_steps": 150, "total_steps": 190, "eval_loss": 0.42412105202674866, "epoch": 3.966442953020134, "percentage": 78.95, "elapsed_time": "1:10:31", "remaining_time": "0:18:48", "throughput": 17795.72, "total_tokens": 75296576}
+{"current_steps": 151, "total_steps": 190, "loss": 1.9097, "lr": 6.452143679117964e-06, "epoch": 3.9932885906040267, "percentage": 79.47, "elapsed_time": "1:11:57", "remaining_time": "0:18:35", "throughput": 17557.68, "total_tokens": 75810848}
+{"current_steps": 152, "total_steps": 190, "loss": 0.2221, "lr": 6.147334755577596e-06, "epoch": 4.0, "percentage": 80.0, "elapsed_time": "1:12:08", "remaining_time": "0:18:02", "throughput": 17543.09, "total_tokens": 75940992}
+{"current_steps": 153, "total_steps": 190, "loss": 0.2591, "lr": 5.848888922025553e-06, "epoch": 4.026845637583893, "percentage": 80.53, "elapsed_time": "1:12:34", "remaining_time": "0:17:33", "throughput": 17557.45, "total_tokens": 76456096}
+{"current_steps": 154, "total_steps": 190, "loss": 0.2145, "lr": 5.556906908924655e-06, "epoch": 4.053691275167785, "percentage": 81.05, "elapsed_time": "1:13:00", "remaining_time": "0:17:04", "throughput": 17572.03, "total_tokens": 76975488}
+{"current_steps": 155, "total_steps": 190, "loss": 0.212, "lr": 5.271487265090163e-06, "epoch": 4.080536912751678, "percentage": 81.58, "elapsed_time": "1:13:26", "remaining_time": "0:16:35", "throughput": 17583.51, "total_tokens": 77487072}
+{"current_steps": 156, "total_steps": 190, "loss": 0.267, "lr": 4.992726324427901e-06, "epoch": 4.10738255033557, "percentage": 82.11, "elapsed_time": "1:13:52", "remaining_time": "0:16:06", "throughput": 17598.3, "total_tokens": 78007040}
+{"current_steps": 157, "total_steps": 190, "loss": 0.1877, "lr": 4.720718173419947e-06, "epoch": 4.134228187919463, "percentage": 82.63, "elapsed_time": "1:14:18", "remaining_time": "0:15:37", "throughput": 17607.76, "total_tokens": 78506528}
+{"current_steps": 158, "total_steps": 190, "loss": 0.5, "lr": 4.4555546193688735e-06, "epoch": 4.1610738255033555, "percentage": 83.16, "elapsed_time": "1:14:45", "remaining_time": "0:15:08", "throughput": 17618.5, "total_tokens": 79019872}
+{"current_steps": 159, "total_steps": 190, "loss": 0.3085, "lr": 4.19732515941125e-06, "epoch": 4.1879194630872485, "percentage": 83.68, "elapsed_time": "1:15:11", "remaining_time": "0:14:39", "throughput": 17629.39, "total_tokens": 79531008}
+{"current_steps": 160, "total_steps": 190, "loss": 0.2403, "lr": 3.94611695031086e-06, "epoch": 4.214765100671141, "percentage": 84.21, "elapsed_time": "1:15:37", "remaining_time": "0:14:10", "throughput": 17638.75, "total_tokens": 80040288}
+{"current_steps": 161, "total_steps": 190, "loss": 0.1888, "lr": 3.7020147790418263e-06, "epoch": 4.241610738255034, "percentage": 84.74, "elapsed_time": "1:16:03", "remaining_time": "0:13:41", "throughput": 17649.33, "total_tokens": 80541216}
+{"current_steps": 162, "total_steps": 190, "loss": 0.2003, "lr": 3.4651010341716028e-06, "epoch": 4.268456375838926, "percentage": 85.26, "elapsed_time": "1:16:29", "remaining_time": "0:13:13", "throughput": 17661.54, "total_tokens": 81055936}
+{"current_steps": 163, "total_steps": 190, "loss": 1.3116, "lr": 3.2354556780534424e-06, "epoch": 4.295302013422819, "percentage": 85.79, "elapsed_time": "1:16:55", "remaining_time": "0:12:44", "throughput": 17673.39, "total_tokens": 81567328}
+{"current_steps": 164, "total_steps": 190, "loss": 0.3799, "lr": 3.013156219837776e-06, "epoch": 4.322147651006711, "percentage": 86.32, "elapsed_time": "1:17:21", "remaining_time": "0:12:15", "throughput": 17683.66, "total_tokens": 82075904}
+{"current_steps": 165, "total_steps": 190, "loss": 0.8396, "lr": 2.7982776893115627e-06, "epoch": 4.348993288590604, "percentage": 86.84, "elapsed_time": "1:17:46", "remaining_time": "0:11:47", "throughput": 17694.65, "total_tokens": 82579648}
+{"current_steps": 166, "total_steps": 190, "loss": 0.2308, "lr": 2.5908926115744997e-06, "epoch": 4.375838926174497, "percentage": 87.37, "elapsed_time": "1:18:12", "remaining_time": "0:11:18", "throughput": 17705.11, "total_tokens": 83089792}
+{"current_steps": 167, "total_steps": 190, "loss": 0.2354, "lr": 2.391070982560564e-06, "epoch": 4.402684563758389, "percentage": 87.89, "elapsed_time": "1:18:38", "remaining_time": "0:10:49", "throughput": 17710.67, "total_tokens": 83575232}
+{"current_steps": 168, "total_steps": 190, "loss": 0.1976, "lr": 2.1988802454132403e-06, "epoch": 4.429530201342282, "percentage": 88.42, "elapsed_time": "1:19:05", "remaining_time": "0:10:21", "throughput": 17720.1, "total_tokens": 84087168}
+{"current_steps": 169, "total_steps": 190, "loss": 0.1741, "lr": 2.0143852677223075e-06, "epoch": 4.456375838926174, "percentage": 88.95, "elapsed_time": "1:19:31", "remaining_time": "0:09:52", "throughput": 17730.41, "total_tokens": 84598784}
+{"current_steps": 170, "total_steps": 190, "loss": 0.6518, "lr": 1.837648319629956e-06, "epoch": 4.483221476510067, "percentage": 89.47, "elapsed_time": "1:19:57", "remaining_time": "0:09:24", "throughput": 17742.79, "total_tokens": 85116288}
+{"current_steps": 171, "total_steps": 190, "loss": 0.2624, "lr": 1.6687290528135723e-06, "epoch": 4.510067114093959, "percentage": 90.0, "elapsed_time": "1:20:23", "remaining_time": "0:08:55", "throughput": 17753.59, "total_tokens": 85632928}
+{"current_steps": 172, "total_steps": 190, "loss": 0.27, "lr": 1.5076844803522922e-06, "epoch": 4.5369127516778525, "percentage": 90.53, "elapsed_time": "1:20:49", "remaining_time": "0:08:27", "throughput": 17764.62, "total_tokens": 86154432}
+{"current_steps": 173, "total_steps": 190, "loss": 0.2335, "lr": 1.3545689574841342e-06, "epoch": 4.563758389261745, "percentage": 91.05, "elapsed_time": "1:21:15", "remaining_time": "0:07:59", "throughput": 17774.88, "total_tokens": 86669984}
+{"current_steps": 174, "total_steps": 190, "loss": 0.2193, "lr": 1.2094341632602064e-06, "epoch": 4.590604026845638, "percentage": 91.58, "elapsed_time": "1:21:42", "remaining_time": "0:07:30", "throughput": 17783.21, "total_tokens": 87174880}
+{"current_steps": 175, "total_steps": 190, "loss": 0.6403, "lr": 1.0723290831021471e-06, "epoch": 4.617449664429531, "percentage": 92.11, "elapsed_time": "1:22:08", "remaining_time": "0:07:02", "throughput": 17794.12, "total_tokens": 87699168}
+{"current_steps": 175, "total_steps": 190, "eval_loss": 0.4115161895751953, "epoch": 4.617449664429531, "percentage": 92.11, "elapsed_time": "1:22:18", "remaining_time": "0:07:03", "throughput": 17757.9, "total_tokens": 87699168}
+{"current_steps": 176, "total_steps": 190, "loss": 0.3758, "lr": 9.432999922687396e-07, "epoch": 4.644295302013423, "percentage": 92.63, "elapsed_time": "1:23:45", "remaining_time": "0:06:39", "throughput": 17554.24, "total_tokens": 88223360}
+{"current_steps": 177, "total_steps": 190, "loss": 0.2342, "lr": 8.223904402372334e-07, "epoch": 4.671140939597316, "percentage": 93.16, "elapsed_time": "1:24:12", "remaining_time": "0:06:11", "throughput": 17565.85, "total_tokens": 88746912}
+{"current_steps": 178, "total_steps": 190, "loss": 0.2333, "lr": 7.096412360046545e-07, "epoch": 4.697986577181208, "percentage": 93.68, "elapsed_time": "1:24:38", "remaining_time": "0:05:42", "throughput": 17576.73, "total_tokens": 89264288}
+{"current_steps": 179, "total_steps": 190, "loss": 0.4271, "lr": 6.050904343141095e-07, "epoch": 4.724832214765101, "percentage": 94.21, "elapsed_time": "1:25:04", "remaining_time": "0:05:13", "throughput": 17587.8, "total_tokens": 89770272}
+{"current_steps": 180, "total_steps": 190, "loss": 0.3234, "lr": 5.087733228106517e-07, "epoch": 4.751677852348993, "percentage": 94.74, "elapsed_time": "1:25:30", "remaining_time": "0:04:45", "throughput": 17599.28, "total_tokens": 90293344}
+{"current_steps": 181, "total_steps": 190, "loss": 0.2556, "lr": 4.207224101311247e-07, "epoch": 4.778523489932886, "percentage": 95.26, "elapsed_time": "1:25:56", "remaining_time": "0:04:16", "throughput": 17610.73, "total_tokens": 90813536}
+{"current_steps": 182, "total_steps": 190, "loss": 0.1983, "lr": 3.4096741493194197e-07, "epoch": 4.805369127516778, "percentage": 95.79, "elapsed_time": "1:26:22", "remaining_time": "0:03:47", "throughput": 17618.27, "total_tokens": 91300416}
+{"current_steps": 183, "total_steps": 190, "loss": 0.2342, "lr": 2.6953525585855234e-07, "epoch": 4.832214765100671, "percentage": 96.32, "elapsed_time": "1:26:47", "remaining_time": "0:03:19", "throughput": 17630.24, "total_tokens": 91817952}
+{"current_steps": 184, "total_steps": 190, "loss": 0.2582, "lr": 2.064500424599436e-07, "epoch": 4.859060402684563, "percentage": 96.84, "elapsed_time": "1:27:14", "remaining_time": "0:02:50", "throughput": 17640.04, "total_tokens": 92328800}
+{"current_steps": 185, "total_steps": 190, "loss": 0.2364, "lr": 1.517330670512629e-07, "epoch": 4.885906040268456, "percentage": 97.37, "elapsed_time": "1:27:40", "remaining_time": "0:02:22", "throughput": 17650.03, "total_tokens": 92843328}
+{"current_steps": 186, "total_steps": 190, "loss": 0.1979, "lr": 1.0540279752731253e-07, "epoch": 4.912751677852349, "percentage": 97.89, "elapsed_time": "1:28:06", "remaining_time": "0:01:53", "throughput": 17656.46, "total_tokens": 93340480}
+{"current_steps": 187, "total_steps": 190, "loss": 0.6738, "lr": 6.747487112931661e-08, "epoch": 4.939597315436242, "percentage": 98.42, "elapsed_time": "1:28:33", "remaining_time": "0:01:25", "throughput": 17664.53, "total_tokens": 93856128}
+{"current_steps": 188, "total_steps": 190, "loss": 0.1781, "lr": 3.796208916709565e-08, "epoch": 4.966442953020135, "percentage": 98.95, "elapsed_time": "1:28:59", "remaining_time": "0:00:56", "throughput": 17674.52, "total_tokens": 94377760}
+{"current_steps": 189, "total_steps": 190, "loss": 0.2207, "lr": 1.6874412698408836e-08, "epoch": 4.993288590604027, "percentage": 99.47, "elapsed_time": "1:29:26", "remaining_time": "0:00:28", "throughput": 17683.04, "total_tokens": 94890496}
+{"current_steps": 190, "total_steps": 190, "loss": 0.3254, "lr": 4.218959166932268e-09, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "1:29:31", "remaining_time": "0:00:00", "throughput": 17690.65, "total_tokens": 95021568}
+{"current_steps": 190, "total_steps": 190, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "1:30:31", "remaining_time": "0:00:00", "throughput": 17493.25, "total_tokens": 95021568}

result_model/trainer_state.json

@@ -0,0 +1,2014 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 25,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.026845637583892617,
+      "grad_norm": 8.237038612365723,
+      "learning_rate": 0.0,
+      "loss": 0.5587,
+      "num_input_tokens_seen": 496000,
+      "step": 1,
+      "train_runtime": 35.4925,
+      "train_tokens_per_second": 13974.782
+    },
+    {
+      "epoch": 0.053691275167785234,
+      "grad_norm": 8.948539733886719,
+      "learning_rate": 2.631578947368421e-06,
+      "loss": 0.6532,
+      "num_input_tokens_seen": 1007328,
+      "step": 2,
+      "train_runtime": 61.7059,
+      "train_tokens_per_second": 16324.653
+    },
+    {
+      "epoch": 0.08053691275167785,
+      "grad_norm": 5.373775482177734,
+      "learning_rate": 5.263157894736842e-06,
+      "loss": 0.4746,
+      "num_input_tokens_seen": 1520160,
+      "step": 3,
+      "train_runtime": 87.5894,
+      "train_tokens_per_second": 17355.526
+    },
+    {
+      "epoch": 0.10738255033557047,
+      "grad_norm": 8.458134651184082,
+      "learning_rate": 7.894736842105263e-06,
+      "loss": 0.5696,
+      "num_input_tokens_seen": 2031424,
+      "step": 4,
+      "train_runtime": 113.5046,
+      "train_tokens_per_second": 17897.28
+    },
+    {
+      "epoch": 0.1342281879194631,
+      "grad_norm": 7.278451442718506,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.5954,
+      "num_input_tokens_seen": 2542624,
+      "step": 5,
+      "train_runtime": 139.4544,
+      "train_tokens_per_second": 18232.649
+    },
+    {
+      "epoch": 0.1610738255033557,
+      "grad_norm": 5.226474285125732,
+      "learning_rate": 1.3157894736842106e-05,
+      "loss": 0.7396,
+      "num_input_tokens_seen": 3056000,
+      "step": 6,
+      "train_runtime": 165.2601,
+      "train_tokens_per_second": 18492.064
+    },
+    {
+      "epoch": 0.18791946308724833,
+      "grad_norm": 5.280333995819092,
+      "learning_rate": 1.5789473684210526e-05,
+      "loss": 0.7118,
+      "num_input_tokens_seen": 3574176,
+      "step": 7,
+      "train_runtime": 191.221,
+      "train_tokens_per_second": 18691.335
+    },
+    {
+      "epoch": 0.21476510067114093,
+      "grad_norm": 3.394676685333252,
+      "learning_rate": 1.8421052631578947e-05,
+      "loss": 0.5509,
+      "num_input_tokens_seen": 4076224,
+      "step": 8,
+      "train_runtime": 217.8993,
+      "train_tokens_per_second": 18706.918
+    },
+    {
+      "epoch": 0.24161073825503357,
+      "grad_norm": 2.9875869750976562,
+      "learning_rate": 2.105263157894737e-05,
+      "loss": 0.9419,
+      "num_input_tokens_seen": 4582944,
+      "step": 9,
+      "train_runtime": 243.8647,
+      "train_tokens_per_second": 18792.981
+    },
+    {
+      "epoch": 0.2684563758389262,
+      "grad_norm": 9.175309181213379,
+      "learning_rate": 2.368421052631579e-05,
+      "loss": 0.6846,
+      "num_input_tokens_seen": 5083968,
+      "step": 10,
+      "train_runtime": 270.1354,
+      "train_tokens_per_second": 18820.07
+    },
+    {
+      "epoch": 0.2953020134228188,
+      "grad_norm": 6.438014507293701,
+      "learning_rate": 2.6315789473684212e-05,
+      "loss": 0.5596,
+      "num_input_tokens_seen": 5608256,
+      "step": 11,
+      "train_runtime": 296.43,
+      "train_tokens_per_second": 18919.326
+    },
+    {
+      "epoch": 0.3221476510067114,
+      "grad_norm": 3.9598963260650635,
+      "learning_rate": 2.8947368421052634e-05,
+      "loss": 0.5313,
+      "num_input_tokens_seen": 6113696,
+      "step": 12,
+      "train_runtime": 322.4122,
+      "train_tokens_per_second": 18962.36
+    },
+    {
+      "epoch": 0.348993288590604,
+      "grad_norm": 4.5351409912109375,
+      "learning_rate": 3.157894736842105e-05,
+      "loss": 0.7476,
+      "num_input_tokens_seen": 6599840,
+      "step": 13,
+      "train_runtime": 348.2235,
+      "train_tokens_per_second": 18952.887
+    },
+    {
+      "epoch": 0.37583892617449666,
+      "grad_norm": 3.2552919387817383,
+      "learning_rate": 3.421052631578947e-05,
+      "loss": 0.6272,
+      "num_input_tokens_seen": 7116640,
+      "step": 14,
+      "train_runtime": 374.5779,
+      "train_tokens_per_second": 18999.09
+    },
+    {
+      "epoch": 0.40268456375838924,
+      "grad_norm": 3.1868414878845215,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 0.4918,
+      "num_input_tokens_seen": 7631552,
+      "step": 15,
+      "train_runtime": 400.4733,
+      "train_tokens_per_second": 19056.334
+    },
+    {
+      "epoch": 0.42953020134228187,
+      "grad_norm": 3.6740667819976807,
+      "learning_rate": 3.9473684210526316e-05,
+      "loss": 0.5583,
+      "num_input_tokens_seen": 8129664,
+      "step": 16,
+      "train_runtime": 426.6064,
+      "train_tokens_per_second": 19056.59
+    },
+    {
+      "epoch": 0.4563758389261745,
+      "grad_norm": 2.7044334411621094,
+      "learning_rate": 4.210526315789474e-05,
+      "loss": 0.6327,
+      "num_input_tokens_seen": 8648320,
+      "step": 17,
+      "train_runtime": 452.6115,
+      "train_tokens_per_second": 19107.599
+    },
+    {
+      "epoch": 0.48322147651006714,
+      "grad_norm": 2.975066661834717,
+      "learning_rate": 4.473684210526316e-05,
+      "loss": 0.5684,
+      "num_input_tokens_seen": 9166688,
+      "step": 18,
+      "train_runtime": 478.7852,
+      "train_tokens_per_second": 19145.721
+    },
+    {
+      "epoch": 0.5100671140939598,
+      "grad_norm": 4.676921367645264,
+      "learning_rate": 4.736842105263158e-05,
+      "loss": 0.6136,
+      "num_input_tokens_seen": 9689760,
+      "step": 19,
+      "train_runtime": 504.6638,
+      "train_tokens_per_second": 19200.427
+    },
+    {
+      "epoch": 0.5369127516778524,
+      "grad_norm": 3.6379780769348145,
+      "learning_rate": 5e-05,
+      "loss": 0.6268,
+      "num_input_tokens_seen": 10202784,
+      "step": 20,
+      "train_runtime": 530.9453,
+      "train_tokens_per_second": 19216.263
+    },
+    {
+      "epoch": 0.5637583892617449,
+      "grad_norm": 3.534153699874878,
+      "learning_rate": 4.999578104083307e-05,
+      "loss": 0.6514,
+      "num_input_tokens_seen": 10704736,
+      "step": 21,
+      "train_runtime": 556.8939,
+      "train_tokens_per_second": 19222.219
+    },
+    {
+      "epoch": 0.5906040268456376,
+      "grad_norm": 2.571445941925049,
+      "learning_rate": 4.998312558730159e-05,
+      "loss": 0.4698,
+      "num_input_tokens_seen": 11204608,
+      "step": 22,
+      "train_runtime": 583.084,
+      "train_tokens_per_second": 19216.113
+    },
+    {
+      "epoch": 0.6174496644295302,
+      "grad_norm": 4.351921558380127,
+      "learning_rate": 4.996203791083291e-05,
+      "loss": 0.6816,
+      "num_input_tokens_seen": 11702432,
+      "step": 23,
+      "train_runtime": 609.1711,
+      "train_tokens_per_second": 19210.418
+    },
+    {
+      "epoch": 0.6442953020134228,
+      "grad_norm": 2.935748338699341,
+      "learning_rate": 4.993252512887069e-05,
+      "loss": 0.6014,
+      "num_input_tokens_seen": 12210336,
+      "step": 24,
+      "train_runtime": 635.5688,
+      "train_tokens_per_second": 19211.666
+    },
+    {
+      "epoch": 0.6711409395973155,
+      "grad_norm": 10.634203910827637,
+      "learning_rate": 4.9894597202472696e-05,
+      "loss": 2.6525,
+      "num_input_tokens_seen": 12725728,
+      "step": 25,
+      "train_runtime": 661.4676,
+      "train_tokens_per_second": 19238.625
+    },
+    {
+      "epoch": 0.6711409395973155,
+      "eval_accuracy": 0.7960947463461368,
+      "eval_loss": 0.7538173198699951,
+      "eval_runtime": 10.0636,
+      "eval_samples_per_second": 12.421,
+      "eval_steps_per_second": 0.795,
+      "num_input_tokens_seen": 12725728,
+      "step": 25
+    },
+    {
+      "epoch": 0.697986577181208,
+      "grad_norm": 187.8636016845703,
+      "learning_rate": 4.984826693294874e-05,
+      "loss": 3.0007,
+      "num_input_tokens_seen": 13241920,
+      "step": 26,
+      "train_runtime": 762.7143,
+      "train_tokens_per_second": 17361.574
+    },
+    {
+      "epoch": 0.7248322147651006,
+      "grad_norm": 14.758489608764648,
+      "learning_rate": 4.979354995754006e-05,
+      "loss": 1.1118,
+      "num_input_tokens_seen": 13754432,
+      "step": 27,
+      "train_runtime": 788.7937,
+      "train_tokens_per_second": 17437.3
+    },
+    {
+      "epoch": 0.7516778523489933,
+      "grad_norm": 137.02674865722656,
+      "learning_rate": 4.9730464744141445e-05,
+      "loss": 5.0379,
+      "num_input_tokens_seen": 14269824,
+      "step": 28,
+      "train_runtime": 814.7738,
+      "train_tokens_per_second": 17513.848
+    },
+    {
+      "epoch": 0.7785234899328859,
+      "grad_norm": 36.753379821777344,
+      "learning_rate": 4.965903258506806e-05,
+      "loss": 1.1003,
+      "num_input_tokens_seen": 14756000,
+      "step": 29,
+      "train_runtime": 840.804,
+      "train_tokens_per_second": 17549.868
+    },
+    {
+      "epoch": 0.8053691275167785,
+      "grad_norm": 31.222320556640625,
+      "learning_rate": 4.957927758986888e-05,
+      "loss": 1.1593,
+      "num_input_tokens_seen": 15256512,
+      "step": 30,
+      "train_runtime": 867.0437,
+      "train_tokens_per_second": 17596.012
+    },
+    {
+      "epoch": 0.8322147651006712,
+      "grad_norm": 6.577232837677002,
+      "learning_rate": 4.949122667718935e-05,
+      "loss": 0.6791,
+      "num_input_tokens_seen": 15778720,
+      "step": 31,
+      "train_runtime": 892.9916,
+      "train_tokens_per_second": 17669.506
+    },
+    {
+      "epoch": 0.8590604026845637,
+      "grad_norm": 2.800877809524536,
+      "learning_rate": 4.9394909565685894e-05,
+      "loss": 0.8403,
+      "num_input_tokens_seen": 16291104,
+      "step": 32,
+      "train_runtime": 918.733,
+      "train_tokens_per_second": 17732.141
+    },
+    {
+      "epoch": 0.8859060402684564,
+      "grad_norm": 4.291952610015869,
+      "learning_rate": 4.929035876399535e-05,
+      "loss": 0.6432,
+      "num_input_tokens_seen": 16811968,
+      "step": 33,
+      "train_runtime": 944.6114,
+      "train_tokens_per_second": 17797.761
+    },
+    {
+      "epoch": 0.912751677852349,
+      "grad_norm": 2.574287176132202,
+      "learning_rate": 4.917760955976277e-05,
+      "loss": 0.5564,
+      "num_input_tokens_seen": 17327040,
+      "step": 34,
+      "train_runtime": 970.7914,
+      "train_tokens_per_second": 17848.366
+    },
+    {
+      "epoch": 0.9395973154362416,
+      "grad_norm": 57.77891159057617,
+      "learning_rate": 4.905670000773126e-05,
+      "loss": 1.8105,
+      "num_input_tokens_seen": 17823168,
+      "step": 35,
+      "train_runtime": 996.8832,
+      "train_tokens_per_second": 17878.892
+    },
+    {
+      "epoch": 0.9664429530201343,
+      "grad_norm": 6.420737266540527,
+      "learning_rate": 4.892767091689786e-05,
+      "loss": 0.9537,
+      "num_input_tokens_seen": 18343840,
+      "step": 36,
+      "train_runtime": 1023.1655,
+      "train_tokens_per_second": 17928.516
+    },
+    {
+      "epoch": 0.9932885906040269,
+      "grad_norm": 3.768702507019043,
+      "learning_rate": 4.87905658367398e-05,
+      "loss": 0.7753,
+      "num_input_tokens_seen": 18864096,
+      "step": 37,
+      "train_runtime": 1049.136,
+      "train_tokens_per_second": 17980.601
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 3.768702507019043,
+      "learning_rate": 4.864543104251587e-05,
+      "loss": 0.3534,
+      "num_input_tokens_seen": 18979328,
+      "step": 38,
+      "train_runtime": 1054.127,
+      "train_tokens_per_second": 18004.784
+    },
+    {
+      "epoch": 1.0268456375838926,
+      "grad_norm": 7.989157199859619,
+      "learning_rate": 4.849231551964771e-05,
+      "loss": 1.6501,
+      "num_input_tokens_seen": 19477504,
+      "step": 39,
+      "train_runtime": 1080.4599,
+      "train_tokens_per_second": 18027.05
+    },
+    {
+      "epoch": 1.0536912751677852,
+      "grad_norm": 3.0987582206726074,
+      "learning_rate": 4.833127094718643e-05,
+      "loss": 2.3538,
+      "num_input_tokens_seen": 19985312,
+      "step": 40,
+      "train_runtime": 1106.2719,
+      "train_tokens_per_second": 18065.462
+    },
+    {
+      "epoch": 1.0805369127516777,
+      "grad_norm": 2.594900608062744,
+      "learning_rate": 4.8162351680370044e-05,
+      "loss": 0.5087,
+      "num_input_tokens_seen": 20476800,
+      "step": 41,
+      "train_runtime": 1132.3862,
+      "train_tokens_per_second": 18082.877
+    },
+    {
+      "epoch": 1.1073825503355705,
+      "grad_norm": 3.6039469242095947,
+      "learning_rate": 4.79856147322777e-05,
+      "loss": 0.5662,
+      "num_input_tokens_seen": 20980288,
+      "step": 42,
+      "train_runtime": 1158.4548,
+      "train_tokens_per_second": 18110.58
+    },
+    {
+      "epoch": 1.1342281879194631,
+      "grad_norm": 2.1966166496276855,
+      "learning_rate": 4.7801119754586766e-05,
+      "loss": 0.4659,
+      "num_input_tokens_seen": 21492736,
+      "step": 43,
+      "train_runtime": 1184.45,
+      "train_tokens_per_second": 18145.752
+    },
+    {
+      "epoch": 1.1610738255033557,
+      "grad_norm": 2.154080629348755,
+      "learning_rate": 4.760892901743944e-05,
+      "loss": 1.9776,
+      "num_input_tokens_seen": 21997856,
+      "step": 44,
+      "train_runtime": 1210.4933,
+      "train_tokens_per_second": 18172.637
+    },
+    {
+      "epoch": 1.1879194630872483,
+      "grad_norm": 68.91860961914062,
+      "learning_rate": 4.7409107388425504e-05,
+      "loss": 3.0839,
+      "num_input_tokens_seen": 22509088,
+      "step": 45,
+      "train_runtime": 1236.7017,
+      "train_tokens_per_second": 18200.903
+    },
+    {
+      "epoch": 1.2147651006711409,
+      "grad_norm": 24.943208694458008,
+      "learning_rate": 4.7201722310688445e-05,
+      "loss": 0.6674,
+      "num_input_tokens_seen": 23024384,
+      "step": 46,
+      "train_runtime": 1263.3678,
+      "train_tokens_per_second": 18224.609
+    },
+    {
+      "epoch": 1.2416107382550337,
+      "grad_norm": 5.5658488273620605,
+      "learning_rate": 4.698684378016222e-05,
+      "loss": 1.0302,
+      "num_input_tokens_seen": 23536384,
+      "step": 47,
+      "train_runtime": 1289.0088,
+      "train_tokens_per_second": 18259.289
+    },
+    {
+      "epoch": 1.2684563758389262,
+      "grad_norm": 34.10597229003906,
+      "learning_rate": 4.676454432194656e-05,
+      "loss": 1.3633,
+      "num_input_tokens_seen": 24046432,
+      "step": 48,
+      "train_runtime": 1315.3663,
+      "train_tokens_per_second": 18281.167
+    },
+    {
+      "epoch": 1.2953020134228188,
+      "grad_norm": 3.001063585281372,
+      "learning_rate": 4.6534898965828405e-05,
+      "loss": 0.5103,
+      "num_input_tokens_seen": 24549600,
+      "step": 49,
+      "train_runtime": 1341.5667,
+      "train_tokens_per_second": 18299.203
+    },
+    {
+      "epoch": 1.3221476510067114,
+      "grad_norm": 2.9493746757507324,
+      "learning_rate": 4.629798522095818e-05,
+      "loss": 0.6796,
+      "num_input_tokens_seen": 25037312,
+      "step": 50,
+      "train_runtime": 1367.4664,
+      "train_tokens_per_second": 18309.27
+    },
+    {
+      "epoch": 1.3221476510067114,
+      "eval_accuracy": 0.8520892900119574,
+      "eval_loss": 0.5981780886650085,
+      "eval_runtime": 10.0754,
+      "eval_samples_per_second": 12.406,
+      "eval_steps_per_second": 0.794,
+      "num_input_tokens_seen": 25037312,
+      "step": 50
+    },
+    {
+      "epoch": 1.348993288590604,
+      "grad_norm": 1.894378423690796,
+      "learning_rate": 4.6053883049689145e-05,
+      "loss": 0.7632,
+      "num_input_tokens_seen": 25548352,
+      "step": 51,
+      "train_runtime": 1462.8406,
+      "train_tokens_per_second": 17464.891
+    },
+    {
+      "epoch": 1.3758389261744965,
+      "grad_norm": 3.0130774974823,
+      "learning_rate": 4.580267484058876e-05,
+      "loss": 0.5686,
+      "num_input_tokens_seen": 26063136,
+      "step": 52,
+      "train_runtime": 1489.1172,
+      "train_tokens_per_second": 17502.407
+    },
+    {
+      "epoch": 1.4026845637583891,
+      "grad_norm": 2.3910205364227295,
+      "learning_rate": 4.554444538063113e-05,
+      "loss": 0.5809,
+      "num_input_tokens_seen": 26574336,
+      "step": 53,
+      "train_runtime": 1515.4499,
+      "train_tokens_per_second": 17535.608
+    },
+    {
+      "epoch": 1.429530201342282,
+      "grad_norm": 2.2223563194274902,
+      "learning_rate": 4.5279281826580056e-05,
+      "loss": 0.4817,
+      "num_input_tokens_seen": 27078496,
+      "step": 54,
+      "train_runtime": 1541.8316,
+      "train_tokens_per_second": 17562.551
+    },
+    {
+      "epoch": 1.4563758389261745,
+      "grad_norm": 1.928465485572815,
+      "learning_rate": 4.5007273675572104e-05,
+      "loss": 0.4111,
+      "num_input_tokens_seen": 27599232,
+      "step": 55,
+      "train_runtime": 1568.1389,
+      "train_tokens_per_second": 17599.992
+    },
+    {
+      "epoch": 1.483221476510067,
+      "grad_norm": 2.8375704288482666,
+      "learning_rate": 4.4728512734909844e-05,
+      "loss": 0.5102,
+      "num_input_tokens_seen": 28099968,
+      "step": 56,
+      "train_runtime": 1594.4215,
+      "train_tokens_per_second": 17623.927
+    },
+    {
+      "epoch": 1.5100671140939599,
+      "grad_norm": 2.123541831970215,
+      "learning_rate": 4.444309309107535e-05,
+      "loss": 0.8591,
+      "num_input_tokens_seen": 28604864,
+      "step": 57,
+      "train_runtime": 1620.7434,
+      "train_tokens_per_second": 17649.224
+    },
+    {
+      "epoch": 1.5369127516778525,
+      "grad_norm": 7.7735066413879395,
+      "learning_rate": 4.415111107797445e-05,
+      "loss": 0.6124,
+      "num_input_tokens_seen": 29112480,
+      "step": 58,
+      "train_runtime": 1646.8857,
+      "train_tokens_per_second": 17677.293
+    },
+    {
+      "epoch": 1.563758389261745,
+      "grad_norm": 51.44823455810547,
+      "learning_rate": 4.385266524442241e-05,
+      "loss": 0.8082,
+      "num_input_tokens_seen": 29610208,
+      "step": 59,
+      "train_runtime": 1672.6373,
+      "train_tokens_per_second": 17702.707
+    },
+    {
+      "epoch": 1.5906040268456376,
+      "grad_norm": 2.954522132873535,
+      "learning_rate": 4.3547856320882044e-05,
+      "loss": 0.5294,
+      "num_input_tokens_seen": 30121920,
+      "step": 60,
+      "train_runtime": 1698.4583,
+      "train_tokens_per_second": 17734.86
+    },
+    {
+      "epoch": 1.6174496644295302,
+      "grad_norm": 2.5693578720092773,
+      "learning_rate": 4.3236787185465525e-05,
+      "loss": 0.5192,
+      "num_input_tokens_seen": 30632928,
+      "step": 61,
+      "train_runtime": 1724.3917,
+      "train_tokens_per_second": 17764.484
+    },
+    {
+      "epoch": 1.6442953020134228,
+      "grad_norm": 4.527597427368164,
+      "learning_rate": 4.2919562829211283e-05,
+      "loss": 0.554,
+      "num_input_tokens_seen": 31139872,
+      "step": 62,
+      "train_runtime": 1750.568,
+      "train_tokens_per_second": 17788.439
+    },
+    {
+      "epoch": 1.6711409395973154,
+      "grad_norm": 1.5587811470031738,
+      "learning_rate": 4.259629032064779e-05,
+      "loss": 0.4231,
+      "num_input_tokens_seen": 31645376,
+      "step": 63,
+      "train_runtime": 1776.4231,
+      "train_tokens_per_second": 17814.098
+    },
+    {
+      "epoch": 1.697986577181208,
+      "grad_norm": 2.3483505249023438,
+      "learning_rate": 4.226707876965611e-05,
+      "loss": 1.475,
+      "num_input_tokens_seen": 32167168,
+      "step": 64,
+      "train_runtime": 1802.9115,
+      "train_tokens_per_second": 17841.79
+    },
+    {
+      "epoch": 1.7248322147651005,
+      "grad_norm": 104.57856750488281,
+      "learning_rate": 4.193203929064353e-05,
+      "loss": 1.6005,
+      "num_input_tokens_seen": 32691456,
+      "step": 65,
+      "train_runtime": 1829.164,
+      "train_tokens_per_second": 17872.348
+    },
+    {
+      "epoch": 1.7516778523489933,
+      "grad_norm": 2.754912853240967,
+      "learning_rate": 4.159128496504053e-05,
+      "loss": 1.1018,
+      "num_input_tokens_seen": 33188672,
+      "step": 66,
+      "train_runtime": 1855.5298,
+      "train_tokens_per_second": 17886.359
+    },
+    {
+      "epoch": 1.778523489932886,
+      "grad_norm": 2.767381429672241,
+      "learning_rate": 4.1244930803134e-05,
+      "loss": 1.2071,
+      "num_input_tokens_seen": 33690880,
+      "step": 67,
+      "train_runtime": 1881.8777,
+      "train_tokens_per_second": 17902.8
+    },
+    {
+      "epoch": 1.8053691275167785,
+      "grad_norm": 2.260093927383423,
+      "learning_rate": 4.089309370524921e-05,
+      "loss": 0.6214,
+      "num_input_tokens_seen": 34196736,
+      "step": 68,
+      "train_runtime": 1907.9434,
+      "train_tokens_per_second": 17923.349
+    },
+    {
+      "epoch": 1.8322147651006713,
+      "grad_norm": 2.873724937438965,
+      "learning_rate": 4.053589242229412e-05,
+      "loss": 0.5573,
+      "num_input_tokens_seen": 34720768,
+      "step": 69,
+      "train_runtime": 1934.0333,
+      "train_tokens_per_second": 17952.518
+    },
+    {
+      "epoch": 1.8590604026845639,
+      "grad_norm": 2.169739007949829,
+      "learning_rate": 4.0173447515678916e-05,
+      "loss": 0.436,
+      "num_input_tokens_seen": 35230112,
+      "step": 70,
+      "train_runtime": 1959.9087,
+      "train_tokens_per_second": 17975.384
+    },
+    {
+      "epoch": 1.8859060402684564,
+      "grad_norm": 2.2907986640930176,
+      "learning_rate": 3.9805881316624506e-05,
+      "loss": 0.4724,
+      "num_input_tokens_seen": 35752064,
+      "step": 71,
+      "train_runtime": 1986.0719,
+      "train_tokens_per_second": 18001.395
+    },
+    {
+      "epoch": 1.912751677852349,
+      "grad_norm": 3.054774284362793,
+      "learning_rate": 3.9433317884873664e-05,
+      "loss": 0.4414,
+      "num_input_tokens_seen": 36274496,
+      "step": 72,
+      "train_runtime": 2012.7145,
+      "train_tokens_per_second": 18022.673
+    },
+    {
+      "epoch": 1.9395973154362416,
+      "grad_norm": 1.7803468704223633,
+      "learning_rate": 3.905588296681886e-05,
+      "loss": 0.3534,
+      "num_input_tokens_seen": 36777152,
+      "step": 73,
+      "train_runtime": 2039.1555,
+      "train_tokens_per_second": 18035.482
+    },
+    {
+      "epoch": 1.9664429530201342,
+      "grad_norm": 111.7915267944336,
+      "learning_rate": 3.867370395306068e-05,
+      "loss": 1.1611,
+      "num_input_tokens_seen": 37298112,
+      "step": 74,
+      "train_runtime": 2065.3775,
+      "train_tokens_per_second": 18058.738
+    },
+    {
+      "epoch": 1.9932885906040267,
+      "grad_norm": 2.8219268321990967,
+      "learning_rate": 3.82869098354114e-05,
+      "loss": 0.6056,
+      "num_input_tokens_seen": 37817120,
+      "step": 75,
+      "train_runtime": 2092.2846,
+      "train_tokens_per_second": 18074.558
+    },
+    {
+      "epoch": 1.9932885906040267,
+      "eval_accuracy": 0.8556461964204233,
+      "eval_loss": 0.5577213764190674,
+      "eval_runtime": 10.0688,
+      "eval_samples_per_second": 12.415,
+      "eval_steps_per_second": 0.795,
+      "num_input_tokens_seen": 37817120,
+      "step": 75
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.8219268321990967,
+      "learning_rate": 3.7895631163358105e-05,
+      "loss": 0.4723,
+      "num_input_tokens_seen": 37948192,
+      "step": 76,
+      "train_runtime": 2167.0471,
+      "train_tokens_per_second": 17511.475
+    },
+    {
+      "epoch": 2.0268456375838926,
+      "grad_norm": 2.0256853103637695,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.4424,
+      "num_input_tokens_seen": 38467968,
+      "step": 77,
+      "train_runtime": 2193.3126,
+      "train_tokens_per_second": 17538.753
+    },
+    {
+      "epoch": 2.053691275167785,
+      "grad_norm": 1.5646251440048218,
+      "learning_rate": 3.7100149877474974e-05,
+      "loss": 1.4039,
+      "num_input_tokens_seen": 38986496,
+      "step": 78,
+      "train_runtime": 2219.7314,
+      "train_tokens_per_second": 17563.609
+    },
+    {
+      "epoch": 2.0805369127516777,
+      "grad_norm": 2.312363386154175,
+      "learning_rate": 3.66962157518902e-05,
+      "loss": 0.3641,
+      "num_input_tokens_seen": 39507008,
+      "step": 79,
+      "train_runtime": 2246.4116,
+      "train_tokens_per_second": 17586.718
+    },
+    {
+      "epoch": 2.1073825503355703,
+      "grad_norm": 1.7589001655578613,
+      "learning_rate": 3.628833395777224e-05,
+      "loss": 0.3558,
+      "num_input_tokens_seen": 40012992,
+      "step": 80,
+      "train_runtime": 2272.6092,
+      "train_tokens_per_second": 17606.632
+    },
+    {
+      "epoch": 2.134228187919463,
+      "grad_norm": 1.479117751121521,
+      "learning_rate": 3.587664216205183e-05,
+      "loss": 0.4281,
+      "num_input_tokens_seen": 40536608,
+      "step": 81,
+      "train_runtime": 2298.9106,
+      "train_tokens_per_second": 17632.964
+    },
+    {
+      "epoch": 2.1610738255033555,
+      "grad_norm": 1.605286717414856,
+      "learning_rate": 3.546127931759903e-05,
+      "loss": 0.3695,
+      "num_input_tokens_seen": 41052192,
+      "step": 82,
+      "train_runtime": 2324.8418,
+      "train_tokens_per_second": 17658.058
+    },
+    {
+      "epoch": 2.1879194630872485,
+      "grad_norm": 19.489822387695312,
+      "learning_rate": 3.504238561632424e-05,
+      "loss": 0.967,
+      "num_input_tokens_seen": 41559584,
+      "step": 83,
+      "train_runtime": 2351.0304,
+      "train_tokens_per_second": 17677.179
+    },
+    {
+      "epoch": 2.214765100671141,
+      "grad_norm": 3.9726529121398926,
+      "learning_rate": 3.4620102441861143e-05,
+      "loss": 0.4472,
+      "num_input_tokens_seen": 42060128,
+      "step": 84,
+      "train_runtime": 2377.0613,
+      "train_tokens_per_second": 17694.171
+    },
+    {
+      "epoch": 2.2416107382550337,
+      "grad_norm": 3.072654962539673,
+      "learning_rate": 3.4194572321847336e-05,
+      "loss": 1.2018,
+      "num_input_tokens_seen": 42570080,
+      "step": 85,
+      "train_runtime": 2403.2948,
+      "train_tokens_per_second": 17713.216
+    },
+    {
+      "epoch": 2.2684563758389262,
+      "grad_norm": 1.8841156959533691,
+      "learning_rate": 3.376593887981887e-05,
+      "loss": 1.1866,
+      "num_input_tokens_seen": 43068832,
+      "step": 86,
+      "train_runtime": 2429.5267,
+      "train_tokens_per_second": 17727.252
+    },
+    {
+      "epoch": 2.295302013422819,
+      "grad_norm": 58.6556396484375,
+      "learning_rate": 3.333434678673489e-05,
+      "loss": 1.9914,
+      "num_input_tokens_seen": 43582016,
+      "step": 87,
+      "train_runtime": 2455.318,
+      "train_tokens_per_second": 17750.049
+    },
+    {
+      "epoch": 2.3221476510067114,
+      "grad_norm": 3.74403715133667,
+      "learning_rate": 3.289994171214882e-05,
+      "loss": 0.6787,
+      "num_input_tokens_seen": 44089344,
+      "step": 88,
+      "train_runtime": 2481.5665,
+      "train_tokens_per_second": 17766.739
+    },
+    {
+      "epoch": 2.348993288590604,
+      "grad_norm": 3.9329142570495605,
+      "learning_rate": 3.246287027504237e-05,
+      "loss": 0.4743,
+      "num_input_tokens_seen": 44589408,
+      "step": 89,
+      "train_runtime": 2507.2669,
+      "train_tokens_per_second": 17784.069
+    },
+    {
+      "epoch": 2.3758389261744965,
+      "grad_norm": 1.8977859020233154,
+      "learning_rate": 3.202327999433924e-05,
+      "loss": 0.4042,
+      "num_input_tokens_seen": 45105888,
+      "step": 90,
+      "train_runtime": 2533.214,
+      "train_tokens_per_second": 17805.795
+    },
+    {
+      "epoch": 2.402684563758389,
+      "grad_norm": 1.4344909191131592,
+      "learning_rate": 3.158131923911498e-05,
+      "loss": 0.4187,
+      "num_input_tokens_seen": 45623968,
+      "step": 91,
+      "train_runtime": 2559.4997,
+      "train_tokens_per_second": 17825.346
+    },
+    {
+      "epoch": 2.4295302013422817,
+      "grad_norm": 2.5110371112823486,
+      "learning_rate": 3.1137137178519985e-05,
+      "loss": 3.5044,
+      "num_input_tokens_seen": 46136544,
+      "step": 92,
+      "train_runtime": 2585.6658,
+      "train_tokens_per_second": 17843.197
+    },
+    {
+      "epoch": 2.4563758389261743,
+      "grad_norm": 1.7893449068069458,
+      "learning_rate": 3.069088373143234e-05,
+      "loss": 0.3795,
+      "num_input_tokens_seen": 46651040,
+      "step": 93,
+      "train_runtime": 2611.8151,
+      "train_tokens_per_second": 17861.54
+    },
+    {
+      "epoch": 2.4832214765100673,
+      "grad_norm": 176.00808715820312,
+      "learning_rate": 3.0242709515857758e-05,
+      "loss": 2.3593,
+      "num_input_tokens_seen": 47163616,
+      "step": 94,
+      "train_runtime": 2638.1455,
+      "train_tokens_per_second": 17877.565
+    },
+    {
+      "epoch": 2.51006711409396,
+      "grad_norm": 884.7402954101562,
+      "learning_rate": 2.9792765798093465e-05,
+      "loss": 1.9348,
+      "num_input_tokens_seen": 47674976,
+      "step": 95,
+      "train_runtime": 2663.9769,
+      "train_tokens_per_second": 17896.167
+    },
+    {
+      "epoch": 2.5369127516778525,
+      "grad_norm": 153.87149047851562,
+      "learning_rate": 2.9341204441673266e-05,
+      "loss": 5.973,
+      "num_input_tokens_seen": 48185504,
+      "step": 96,
+      "train_runtime": 2690.5787,
+      "train_tokens_per_second": 17908.974
+    },
+    {
+      "epoch": 2.563758389261745,
+      "grad_norm": 3.753488302230835,
+      "learning_rate": 2.8888177856111083e-05,
+      "loss": 0.5242,
+      "num_input_tokens_seen": 48689920,
+      "step": 97,
+      "train_runtime": 2716.6014,
+      "train_tokens_per_second": 17923.1
+    },
+    {
+      "epoch": 2.5906040268456376,
+      "grad_norm": 4.23100471496582,
+      "learning_rate": 2.8433838945460205e-05,
+      "loss": 0.6778,
+      "num_input_tokens_seen": 49207680,
+      "step": 98,
+      "train_runtime": 2742.4391,
+      "train_tokens_per_second": 17943.034
+    },
+    {
+      "epoch": 2.61744966442953,
+      "grad_norm": 2.3302712440490723,
+      "learning_rate": 2.797834105670559e-05,
+      "loss": 0.4994,
+      "num_input_tokens_seen": 49701344,
+      "step": 99,
+      "train_runtime": 2768.5965,
+      "train_tokens_per_second": 17951.819
+    },
+    {
+      "epoch": 2.6442953020134228,
+      "grad_norm": 1.5278692245483398,
+      "learning_rate": 2.752183792800671e-05,
+      "loss": 0.4019,
+      "num_input_tokens_seen": 50216704,
+      "step": 100,
+      "train_runtime": 2794.7132,
+      "train_tokens_per_second": 17968.464
+    },
+    {
+      "epoch": 2.6442953020134228,
+      "eval_accuracy": 0.868777035442255,
+      "eval_loss": 0.5278336405754089,
+      "eval_runtime": 10.0705,
+      "eval_samples_per_second": 12.412,
+      "eval_steps_per_second": 0.794,
+      "num_input_tokens_seen": 50216704,
+      "step": 100
+    },
+    {
+      "epoch": 2.6711409395973154,
+      "grad_norm": 1.9811456203460693,
+      "learning_rate": 2.7064483636808313e-05,
+      "loss": 1.1764,
+      "num_input_tokens_seen": 50727392,
+      "step": 101,
+      "train_runtime": 2890.7352,
+      "train_tokens_per_second": 17548.267
+    },
+    {
+      "epoch": 2.697986577181208,
+      "grad_norm": 22.440269470214844,
+      "learning_rate": 2.6606432547836757e-05,
+      "loss": 0.8064,
+      "num_input_tokens_seen": 51242496,
+      "step": 102,
+      "train_runtime": 2917.1561,
+      "train_tokens_per_second": 17565.908
+    },
+    {
+      "epoch": 2.7248322147651005,
+      "grad_norm": 1.9680432081222534,
+      "learning_rate": 2.6147839260999295e-05,
+      "loss": 6.5702,
+      "num_input_tokens_seen": 51725728,
+      "step": 103,
+      "train_runtime": 2943.046,
+      "train_tokens_per_second": 17575.575
+    },
+    {
+      "epoch": 2.751677852348993,
+      "grad_norm": 74.59529113769531,
+      "learning_rate": 2.5688858559204053e-05,
+      "loss": 0.3993,
+      "num_input_tokens_seen": 52244192,
+      "step": 104,
+      "train_runtime": 2969.7783,
+      "train_tokens_per_second": 17591.95
+    },
+    {
+      "epoch": 2.778523489932886,
+      "grad_norm": 1.6544747352600098,
+      "learning_rate": 2.5229645356118163e-05,
+      "loss": 0.4268,
+      "num_input_tokens_seen": 52757856,
+      "step": 105,
+      "train_runtime": 2996.1745,
+      "train_tokens_per_second": 17608.405
+    },
+    {
+      "epoch": 2.8053691275167782,
+      "grad_norm": 1.4038758277893066,
+      "learning_rate": 2.4770354643881843e-05,
+      "loss": 0.358,
+      "num_input_tokens_seen": 53267776,
+      "step": 106,
+      "train_runtime": 3022.3818,
+      "train_tokens_per_second": 17624.436
+    },
+    {
+      "epoch": 2.8322147651006713,
+      "grad_norm": 1.7984601259231567,
+      "learning_rate": 2.4311141440795953e-05,
+      "loss": 1.1325,
+      "num_input_tokens_seen": 53774464,
+      "step": 107,
+      "train_runtime": 3048.432,
+      "train_tokens_per_second": 17640.041
+    },
+    {
+      "epoch": 2.859060402684564,
+      "grad_norm": 40.03776931762695,
+      "learning_rate": 2.3852160739000707e-05,
+      "loss": 3.3749,
+      "num_input_tokens_seen": 54287936,
+      "step": 108,
+      "train_runtime": 3074.3539,
+      "train_tokens_per_second": 17658.324
+    },
+    {
+      "epoch": 2.8859060402684564,
+      "grad_norm": 2.0243523120880127,
+      "learning_rate": 2.3393567452163252e-05,
+      "loss": 0.3589,
+      "num_input_tokens_seen": 54776736,
+      "step": 109,
+      "train_runtime": 3100.4425,
+      "train_tokens_per_second": 17667.393
+    },
+    {
+      "epoch": 2.912751677852349,
+      "grad_norm": 1.841299057006836,
+      "learning_rate": 2.2935516363191693e-05,
+      "loss": 0.4193,
+      "num_input_tokens_seen": 55271424,
+      "step": 110,
+      "train_runtime": 3126.5372,
+      "train_tokens_per_second": 17678.16
+    },
+    {
+      "epoch": 2.9395973154362416,
+      "grad_norm": 1.390844702720642,
+      "learning_rate": 2.2478162071993298e-05,
+      "loss": 0.3847,
+      "num_input_tokens_seen": 55769824,
+      "step": 111,
+      "train_runtime": 3152.3296,
+      "train_tokens_per_second": 17691.622
+    },
+    {
+      "epoch": 2.966442953020134,
+      "grad_norm": 1.4262468814849854,
+      "learning_rate": 2.202165894329441e-05,
+      "loss": 0.3429,
+      "num_input_tokens_seen": 56289728,
+      "step": 112,
+      "train_runtime": 3178.39,
+      "train_tokens_per_second": 17710.139
+    },
+    {
+      "epoch": 2.9932885906040267,
+      "grad_norm": 1.5515210628509521,
+      "learning_rate": 2.1566161054539798e-05,
+      "loss": 0.4288,
+      "num_input_tokens_seen": 56806016,
+      "step": 113,
+      "train_runtime": 3204.5586,
+      "train_tokens_per_second": 17726.627
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.5515210628509521,
+      "learning_rate": 2.111182214388893e-05,
+      "loss": 0.389,
+      "num_input_tokens_seen": 56937088,
+      "step": 114,
+      "train_runtime": 3209.7058,
+      "train_tokens_per_second": 17739.036
+    },
+    {
+      "epoch": 3.0268456375838926,
+      "grad_norm": 1.1605178117752075,
+      "learning_rate": 2.0658795558326743e-05,
+      "loss": 3.2378,
+      "num_input_tokens_seen": 57437376,
+      "step": 115,
+      "train_runtime": 3236.2481,
+      "train_tokens_per_second": 17748.137
+    },
+    {
+      "epoch": 3.053691275167785,
+      "grad_norm": 45.809234619140625,
+      "learning_rate": 2.0207234201906547e-05,
+      "loss": 0.7378,
+      "num_input_tokens_seen": 57926240,
+      "step": 116,
+      "train_runtime": 3262.3674,
+      "train_tokens_per_second": 17755.891
+    },
+    {
+      "epoch": 3.0805369127516777,
+      "grad_norm": 44.064762115478516,
+      "learning_rate": 1.9757290484142244e-05,
+      "loss": 0.3071,
+      "num_input_tokens_seen": 58445376,
+      "step": 117,
+      "train_runtime": 3288.5966,
+      "train_tokens_per_second": 17772.133
+    },
+    {
+      "epoch": 3.1073825503355703,
+      "grad_norm": 2.5294344425201416,
+      "learning_rate": 1.9309116268567674e-05,
+      "loss": 0.5286,
+      "num_input_tokens_seen": 58940512,
+      "step": 118,
+      "train_runtime": 3314.2764,
+      "train_tokens_per_second": 17783.825
+    },
+    {
+      "epoch": 3.134228187919463,
+      "grad_norm": 2.2777907848358154,
+      "learning_rate": 1.8862862821480025e-05,
+      "loss": 1.2341,
+      "num_input_tokens_seen": 59447232,
+      "step": 119,
+      "train_runtime": 3340.3057,
+      "train_tokens_per_second": 17796.943
+    },
+    {
+      "epoch": 3.1610738255033555,
+      "grad_norm": 1.2976305484771729,
+      "learning_rate": 1.8418680760885027e-05,
+      "loss": 0.2624,
+      "num_input_tokens_seen": 59962592,
+      "step": 120,
+      "train_runtime": 3366.4974,
+      "train_tokens_per_second": 17811.567
+    },
+    {
+      "epoch": 3.1879194630872485,
+      "grad_norm": 1.4661471843719482,
+      "learning_rate": 1.797672000566077e-05,
+      "loss": 0.3812,
+      "num_input_tokens_seen": 60473632,
+      "step": 121,
+      "train_runtime": 3392.1332,
+      "train_tokens_per_second": 17827.611
+    },
+    {
+      "epoch": 3.214765100671141,
+      "grad_norm": 1.7925889492034912,
+      "learning_rate": 1.7537129724957642e-05,
+      "loss": 0.3298,
+      "num_input_tokens_seen": 60984768,
+      "step": 122,
+      "train_runtime": 3418.1221,
+      "train_tokens_per_second": 17841.6
+    },
+    {
+      "epoch": 3.2416107382550337,
+      "grad_norm": 1.1223887205123901,
+      "learning_rate": 1.710005828785119e-05,
+      "loss": 0.316,
+      "num_input_tokens_seen": 61507680,
+      "step": 123,
+      "train_runtime": 3444.5252,
+      "train_tokens_per_second": 17856.65
+    },
+    {
+      "epoch": 3.2684563758389262,
+      "grad_norm": 1.381909966468811,
+      "learning_rate": 1.666565321326512e-05,
+      "loss": 0.8116,
+      "num_input_tokens_seen": 62023520,
+      "step": 124,
+      "train_runtime": 3470.5519,
+      "train_tokens_per_second": 17871.371
+    },
+    {
+      "epoch": 3.295302013422819,
+      "grad_norm": 40.172916412353516,
+      "learning_rate": 1.6234061120181142e-05,
+      "loss": 0.3133,
+      "num_input_tokens_seen": 62538624,
+      "step": 125,
+      "train_runtime": 3496.8506,
+      "train_tokens_per_second": 17884.271
+    },
+    {
+      "epoch": 3.295302013422819,
+      "eval_accuracy": 0.8777751968802215,
+      "eval_loss": 0.4720778465270996,
+      "eval_runtime": 10.0701,
+      "eval_samples_per_second": 12.413,
+      "eval_steps_per_second": 0.794,
+      "num_input_tokens_seen": 62538624,
+      "step": 125
+    },
+    {
+      "epoch": 3.3221476510067114,
+      "grad_norm": 1.1484184265136719,
+      "learning_rate": 1.5805427678152677e-05,
+      "loss": 0.8765,
+      "num_input_tokens_seen": 63047392,
+      "step": 126,
+      "train_runtime": 3593.249,
+      "train_tokens_per_second": 17546.068
+    },
+    {
+      "epoch": 3.348993288590604,
+      "grad_norm": 0.9760416746139526,
+      "learning_rate": 1.5379897558138862e-05,
+      "loss": 0.2751,
+      "num_input_tokens_seen": 63547936,
+      "step": 127,
+      "train_runtime": 3620.1181,
+      "train_tokens_per_second": 17554.106
+    },
+    {
+      "epoch": 3.3758389261744965,
+      "grad_norm": 41.91545867919922,
+      "learning_rate": 1.495761438367577e-05,
+      "loss": 1.2805,
+      "num_input_tokens_seen": 64051232,
+      "step": 128,
+      "train_runtime": 3646.4211,
+      "train_tokens_per_second": 17565.506
+    },
+    {
+      "epoch": 3.402684563758389,
+      "grad_norm": 1.1068097352981567,
+      "learning_rate": 1.4538720682400969e-05,
+      "loss": 0.2843,
+      "num_input_tokens_seen": 64546048,
+      "step": 129,
+      "train_runtime": 3673.1876,
+      "train_tokens_per_second": 17572.217
+    },
+    {
+      "epoch": 3.4295302013422817,
+      "grad_norm": 1.142320156097412,
+      "learning_rate": 1.4123357837948175e-05,
+      "loss": 0.3334,
+      "num_input_tokens_seen": 65047392,
+      "step": 130,
+      "train_runtime": 3699.4486,
+      "train_tokens_per_second": 17582.997
+    },
+    {
+      "epoch": 3.4563758389261743,
+      "grad_norm": 46.183719635009766,
+      "learning_rate": 1.3711666042227772e-05,
+      "loss": 1.2229,
+      "num_input_tokens_seen": 65565600,
+      "step": 131,
+      "train_runtime": 3725.8008,
+      "train_tokens_per_second": 17597.72
+    },
+    {
+      "epoch": 3.4832214765100673,
+      "grad_norm": 0.8785358667373657,
+      "learning_rate": 1.330378424810981e-05,
+      "loss": 0.4387,
+      "num_input_tokens_seen": 66075392,
+      "step": 132,
+      "train_runtime": 3752.0888,
+      "train_tokens_per_second": 17610.295
+    },
+    {
+      "epoch": 3.51006711409396,
+      "grad_norm": 1.0320525169372559,
+      "learning_rate": 1.2899850122525037e-05,
+      "loss": 0.2675,
+      "num_input_tokens_seen": 66591552,
+      "step": 133,
+      "train_runtime": 3777.8237,
+      "train_tokens_per_second": 17626.961
+    },
+    {
+      "epoch": 3.5369127516778525,
+      "grad_norm": 1.3758822679519653,
+      "learning_rate": 1.2500000000000006e-05,
+      "loss": 0.2816,
+      "num_input_tokens_seen": 67106688,
+      "step": 134,
+      "train_runtime": 3803.8322,
+      "train_tokens_per_second": 17641.863
+    },
+    {
+      "epoch": 3.563758389261745,
+      "grad_norm": 1.0216566324234009,
+      "learning_rate": 1.2104368836641908e-05,
+      "loss": 0.2957,
+      "num_input_tokens_seen": 67618688,
+      "step": 135,
+      "train_runtime": 3829.8553,
+      "train_tokens_per_second": 17655.677
+    },
+    {
+      "epoch": 3.5906040268456376,
+      "grad_norm": 91.42462921142578,
+      "learning_rate": 1.1713090164588607e-05,
+      "loss": 2.8051,
+      "num_input_tokens_seen": 68129952,
+      "step": 136,
+      "train_runtime": 3856.2115,
+      "train_tokens_per_second": 17667.587
+    },
+    {
+      "epoch": 3.61744966442953,
+      "grad_norm": 0.8375159502029419,
+      "learning_rate": 1.1326296046939333e-05,
+      "loss": 0.2361,
+      "num_input_tokens_seen": 68642048,
+      "step": 137,
+      "train_runtime": 3882.2608,
+      "train_tokens_per_second": 17680.947
+    },
+    {
+      "epoch": 3.6442953020134228,
+      "grad_norm": 1.1549395322799683,
+      "learning_rate": 1.0944117033181151e-05,
+      "loss": 0.4118,
+      "num_input_tokens_seen": 69161312,
+      "step": 138,
+      "train_runtime": 3908.295,
+      "train_tokens_per_second": 17696.032
+    },
+    {
+      "epoch": 3.6711409395973154,
+      "grad_norm": 1.7038424015045166,
+      "learning_rate": 1.0566682115126344e-05,
+      "loss": 0.3521,
+      "num_input_tokens_seen": 69669024,
+      "step": 139,
+      "train_runtime": 3934.6575,
+      "train_tokens_per_second": 17706.503
+    },
+    {
+      "epoch": 3.697986577181208,
+      "grad_norm": 0.8558218479156494,
+      "learning_rate": 1.0194118683375503e-05,
+      "loss": 0.3169,
+      "num_input_tokens_seen": 70169152,
+      "step": 140,
+      "train_runtime": 3960.2013,
+      "train_tokens_per_second": 17718.582
+    },
+    {
+      "epoch": 3.7248322147651005,
+      "grad_norm": 1.4320504665374756,
+      "learning_rate": 9.826552484321087e-06,
+      "loss": 0.2943,
+      "num_input_tokens_seen": 70691424,
+      "step": 141,
+      "train_runtime": 3986.6426,
+      "train_tokens_per_second": 17732.069
+    },
+    {
+      "epoch": 3.751677852348993,
+      "grad_norm": 1.188143253326416,
+      "learning_rate": 9.464107577705886e-06,
+      "loss": 0.4079,
+      "num_input_tokens_seen": 71198976,
+      "step": 142,
+      "train_runtime": 4012.8509,
+      "train_tokens_per_second": 17742.742
+    },
+    {
+      "epoch": 3.778523489932886,
+      "grad_norm": 1.1367772817611694,
+      "learning_rate": 9.106906294750805e-06,
+      "loss": 0.2099,
+      "num_input_tokens_seen": 71713312,
+      "step": 143,
+      "train_runtime": 4039.1047,
+      "train_tokens_per_second": 17754.754
+    },
+    {
+      "epoch": 3.8053691275167782,
+      "grad_norm": 0.9287247657775879,
+      "learning_rate": 8.755069196866014e-06,
+      "loss": 0.3496,
+      "num_input_tokens_seen": 72232064,
+      "step": 144,
+      "train_runtime": 4065.3712,
+      "train_tokens_per_second": 17767.643
+    },
+    {
+      "epoch": 3.8322147651006713,
+      "grad_norm": 0.8527315855026245,
+      "learning_rate": 8.40871503495947e-06,
+      "loss": 0.2696,
+      "num_input_tokens_seen": 72743648,
+      "step": 145,
+      "train_runtime": 4091.4699,
+      "train_tokens_per_second": 17779.343
+    },
+    {
+      "epoch": 3.859060402684564,
+      "grad_norm": 1.0350202322006226,
+      "learning_rate": 8.067960709356478e-06,
+      "loss": 0.2302,
+      "num_input_tokens_seen": 73266496,
+      "step": 146,
+      "train_runtime": 4117.9334,
+      "train_tokens_per_second": 17792.055
+    },
+    {
+      "epoch": 3.8859060402684564,
+      "grad_norm": 0.8234533667564392,
+      "learning_rate": 7.732921230343892e-06,
+      "loss": 0.9783,
+      "num_input_tokens_seen": 73779232,
+      "step": 147,
+      "train_runtime": 4144.4299,
+      "train_tokens_per_second": 17802.022
+    },
+    {
+      "epoch": 3.912751677852349,
+      "grad_norm": 1.041733980178833,
+      "learning_rate": 7.403709679352217e-06,
+      "loss": 0.3549,
+      "num_input_tokens_seen": 74283712,
+      "step": 148,
+      "train_runtime": 4170.0695,
+      "train_tokens_per_second": 17813.543
+    },
+    {
+      "epoch": 3.9395973154362416,
+      "grad_norm": 1.2669960260391235,
+      "learning_rate": 7.080437170788723e-06,
+      "loss": 0.303,
+      "num_input_tokens_seen": 74794496,
+      "step": 149,
+      "train_runtime": 4196.4068,
+      "train_tokens_per_second": 17823.462
+    },
+    {
+      "epoch": 3.966442953020134,
+      "grad_norm": 1.1353541612625122,
+      "learning_rate": 6.763212814534484e-06,
+      "loss": 0.3139,
+      "num_input_tokens_seen": 75296576,
+      "step": 150,
+      "train_runtime": 4221.9552,
+      "train_tokens_per_second": 17834.527
+    },
+    {
+      "epoch": 3.966442953020134,
+      "eval_accuracy": 0.8855249776173986,
+      "eval_loss": 0.42412105202674866,
+      "eval_runtime": 10.0868,
+      "eval_samples_per_second": 12.392,
+      "eval_steps_per_second": 0.793,
+      "num_input_tokens_seen": 75296576,
+      "step": 150
+    },
+    {
+      "epoch": 3.9932885906040267,
+      "grad_norm": 0.9825559258460999,
+      "learning_rate": 6.452143679117964e-06,
+      "loss": 1.9097,
+      "num_input_tokens_seen": 75810848,
+      "step": 151,
+      "train_runtime": 4318.6994,
+      "train_tokens_per_second": 17554.092
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.435434103012085,
+      "learning_rate": 6.147334755577596e-06,
+      "loss": 0.2221,
+      "num_input_tokens_seen": 75940992,
+      "step": 152,
+      "train_runtime": 4329.7093,
+      "train_tokens_per_second": 17539.513
+    },
+    {
+      "epoch": 4.026845637583893,
+      "grad_norm": 19.461166381835938,
+      "learning_rate": 5.848888922025553e-06,
+      "loss": 0.2591,
+      "num_input_tokens_seen": 76456096,
+      "step": 153,
+      "train_runtime": 4355.5057,
+      "train_tokens_per_second": 17553.896
+    },
+    {
+      "epoch": 4.053691275167785,
+      "grad_norm": 0.8980227708816528,
+      "learning_rate": 5.556906908924655e-06,
+      "loss": 0.2145,
+      "num_input_tokens_seen": 76975488,
+      "step": 154,
+      "train_runtime": 4381.4504,
+      "train_tokens_per_second": 17568.495
+    },
+    {
+      "epoch": 4.080536912751678,
+      "grad_norm": 0.8387576341629028,
+      "learning_rate": 5.271487265090163e-06,
+      "loss": 0.212,
+      "num_input_tokens_seen": 77487072,
+      "step": 155,
+      "train_runtime": 4407.6872,
+      "train_tokens_per_second": 17579.984
+    },
+    {
+      "epoch": 4.10738255033557,
+      "grad_norm": 0.9640564918518066,
+      "learning_rate": 4.992726324427901e-06,
+      "loss": 0.267,
+      "num_input_tokens_seen": 78007040,
+      "step": 156,
+      "train_runtime": 4433.5277,
+      "train_tokens_per_second": 17594.801
+    },
+    {
+      "epoch": 4.134228187919463,
+      "grad_norm": 0.8446484208106995,
+      "learning_rate": 4.720718173419947e-06,
+      "loss": 0.1877,
+      "num_input_tokens_seen": 78506528,
+      "step": 157,
+      "train_runtime": 4459.5141,
+      "train_tokens_per_second": 17604.279
+    },
+    {
+      "epoch": 4.1610738255033555,
+      "grad_norm": 41.02722930908203,
+      "learning_rate": 4.4555546193688735e-06,
+      "loss": 0.5,
+      "num_input_tokens_seen": 79019872,
+      "step": 158,
+      "train_runtime": 4485.9333,
+      "train_tokens_per_second": 17615.035
+    },
+    {
+      "epoch": 4.1879194630872485,
+      "grad_norm": 5.308436870574951,
+      "learning_rate": 4.19732515941125e-06,
+      "loss": 0.3085,
+      "num_input_tokens_seen": 79531008,
+      "step": 159,
+      "train_runtime": 4512.1558,
+      "train_tokens_per_second": 17625.944
+    },
+    {
+      "epoch": 4.214765100671141,
+      "grad_norm": 1.068027377128601,
+      "learning_rate": 3.94611695031086e-06,
+      "loss": 0.2403,
+      "num_input_tokens_seen": 80040288,
+      "step": 160,
+      "train_runtime": 4538.6362,
+      "train_tokens_per_second": 17635.317
+    },
+    {
+      "epoch": 4.241610738255034,
+      "grad_norm": 0.8074558973312378,
+      "learning_rate": 3.7020147790418263e-06,
+      "loss": 0.1888,
+      "num_input_tokens_seen": 80541216,
+      "step": 161,
+      "train_runtime": 4564.2981,
+      "train_tokens_per_second": 17645.915
+    },
+    {
+      "epoch": 4.268456375838926,
+      "grad_norm": 0.9346213936805725,
+      "learning_rate": 3.4651010341716028e-06,
+      "loss": 0.2003,
+      "num_input_tokens_seen": 81055936,
+      "step": 162,
+      "train_runtime": 4590.2865,
+      "train_tokens_per_second": 17658.143
+    },
+    {
+      "epoch": 4.295302013422819,
+      "grad_norm": 64.37608337402344,
+      "learning_rate": 3.2354556780534424e-06,
+      "loss": 1.3116,
+      "num_input_tokens_seen": 81567328,
+      "step": 163,
+      "train_runtime": 4616.1435,
+      "train_tokens_per_second": 17670.016
+    },
+    {
+      "epoch": 4.322147651006711,
+      "grad_norm": 1.8003010749816895,
+      "learning_rate": 3.013156219837776e-06,
+      "loss": 0.3799,
+      "num_input_tokens_seen": 82075904,
+      "step": 164,
+      "train_runtime": 4642.2255,
+      "train_tokens_per_second": 17680.292
+    },
+    {
+      "epoch": 4.348993288590604,
+      "grad_norm": 51.30678176879883,
+      "learning_rate": 2.7982776893115627e-06,
+      "loss": 0.8396,
+      "num_input_tokens_seen": 82579648,
+      "step": 165,
+      "train_runtime": 4667.8104,
+      "train_tokens_per_second": 17691.303
+    },
+    {
+      "epoch": 4.375838926174497,
+      "grad_norm": 0.9029057621955872,
+      "learning_rate": 2.5908926115744997e-06,
+      "loss": 0.2308,
+      "num_input_tokens_seen": 83089792,
+      "step": 166,
+      "train_runtime": 4693.866,
+      "train_tokens_per_second": 17701.782
+    },
+    {
+      "epoch": 4.402684563758389,
+      "grad_norm": 0.7935118675231934,
+      "learning_rate": 2.391070982560564e-06,
+      "loss": 0.2354,
+      "num_input_tokens_seen": 83575232,
+      "step": 167,
+      "train_runtime": 4719.8017,
+      "train_tokens_per_second": 17707.361
+    },
+    {
+      "epoch": 4.429530201342282,
+      "grad_norm": 0.7607171535491943,
+      "learning_rate": 2.1988802454132403e-06,
+      "loss": 0.1976,
+      "num_input_tokens_seen": 84087168,
+      "step": 168,
+      "train_runtime": 4746.1815,
+      "train_tokens_per_second": 17716.804
+    },
+    {
+      "epoch": 4.456375838926174,
+      "grad_norm": 0.8092939257621765,
+      "learning_rate": 2.0143852677223075e-06,
+      "loss": 0.1741,
+      "num_input_tokens_seen": 84598784,
+      "step": 169,
+      "train_runtime": 4772.2768,
+      "train_tokens_per_second": 17727.133
+    },
+    {
+      "epoch": 4.483221476510067,
+      "grad_norm": 0.8877170085906982,
+      "learning_rate": 1.837648319629956e-06,
+      "loss": 0.6518,
+      "num_input_tokens_seen": 85116288,
+      "step": 170,
+      "train_runtime": 4798.114,
+      "train_tokens_per_second": 17739.53
+    },
+    {
+      "epoch": 4.510067114093959,
+      "grad_norm": 0.7727687358856201,
+      "learning_rate": 1.6687290528135723e-06,
+      "loss": 0.2624,
+      "num_input_tokens_seen": 85632928,
+      "step": 171,
+      "train_runtime": 4824.2965,
+      "train_tokens_per_second": 17750.345
+    },
+    {
+      "epoch": 4.5369127516778525,
+      "grad_norm": 0.933302640914917,
+      "learning_rate": 1.5076844803522922e-06,
+      "loss": 0.27,
+      "num_input_tokens_seen": 86154432,
+      "step": 172,
+      "train_runtime": 4850.6581,
+      "train_tokens_per_second": 17761.39
+    },
+    {
+      "epoch": 4.563758389261745,
+      "grad_norm": 1.2586088180541992,
+      "learning_rate": 1.3545689574841342e-06,
+      "loss": 0.2335,
+      "num_input_tokens_seen": 86669984,
+      "step": 173,
+      "train_runtime": 4876.8647,
+      "train_tokens_per_second": 17771.66
+    },
+    {
+      "epoch": 4.590604026845638,
+      "grad_norm": 0.7908529043197632,
+      "learning_rate": 1.2094341632602064e-06,
+      "loss": 0.2193,
+      "num_input_tokens_seen": 87174880,
+      "step": 174,
+      "train_runtime": 4902.9706,
+      "train_tokens_per_second": 17780.013
+    },
+    {
+      "epoch": 4.617449664429531,
+      "grad_norm": 17.11946678161621,
+      "learning_rate": 1.0723290831021471e-06,
+      "loss": 0.6403,
+      "num_input_tokens_seen": 87699168,
+      "step": 175,
+      "train_runtime": 4929.4301,
+      "train_tokens_per_second": 17790.935
+    },
+    {
+      "epoch": 4.617449664429531,
+      "eval_accuracy": 0.8911537722889794,
+      "eval_loss": 0.4115161895751953,
+      "eval_runtime": 10.0488,
+      "eval_samples_per_second": 12.439,
+      "eval_steps_per_second": 0.796,
+      "num_input_tokens_seen": 87699168,
+      "step": 175
+    },
+    {
+      "epoch": 4.644295302013423,
+      "grad_norm": 2.2624847888946533,
+      "learning_rate": 9.432999922687396e-07,
+      "loss": 0.3758,
+      "num_input_tokens_seen": 88223360,
+      "step": 176,
+      "train_runtime": 5026.639,
+      "train_tokens_per_second": 17551.163
+    },
+    {
+      "epoch": 4.671140939597316,
+      "grad_norm": 0.6959147453308105,
+      "learning_rate": 8.223904402372334e-07,
+      "loss": 0.2342,
+      "num_input_tokens_seen": 88746912,
+      "step": 177,
+      "train_runtime": 5053.1227,
+      "train_tokens_per_second": 17562.786
+    },
+    {
+      "epoch": 4.697986577181208,
+      "grad_norm": 0.5389838814735413,
+      "learning_rate": 7.096412360046545e-07,
+      "loss": 0.2333,
+      "num_input_tokens_seen": 89264288,
+      "step": 178,
+      "train_runtime": 5079.4307,
+      "train_tokens_per_second": 17573.68
+    },
+    {
+      "epoch": 4.724832214765101,
+      "grad_norm": 5.69287109375,
+      "learning_rate": 6.050904343141095e-07,
+      "loss": 0.4271,
+      "num_input_tokens_seen": 89770272,
+      "step": 179,
+      "train_runtime": 5105.0048,
+      "train_tokens_per_second": 17584.758
+    },
+    {
+      "epoch": 4.751677852348993,
+      "grad_norm": 1.6489832401275635,
+      "learning_rate": 5.087733228106517e-07,
+      "loss": 0.3234,
+      "num_input_tokens_seen": 90293344,
+      "step": 180,
+      "train_runtime": 5131.3967,
+      "train_tokens_per_second": 17596.251
+    },
+    {
+      "epoch": 4.778523489932886,
+      "grad_norm": 6.22125768661499,
+      "learning_rate": 4.207224101311247e-07,
+      "loss": 0.2556,
+      "num_input_tokens_seen": 90813536,
+      "step": 181,
+      "train_runtime": 5157.5997,
+      "train_tokens_per_second": 17607.713
+    },
+    {
+      "epoch": 4.805369127516778,
+      "grad_norm": 0.8385538458824158,
+      "learning_rate": 3.4096741493194197e-07,
+      "loss": 0.1983,
+      "num_input_tokens_seen": 91300416,
+      "step": 182,
+      "train_runtime": 5183.0266,
+      "train_tokens_per_second": 17615.271
+    },
+    {
+      "epoch": 4.832214765100671,
+      "grad_norm": 1.0120928287506104,
+      "learning_rate": 2.6953525585855234e-07,
+      "loss": 0.2342,
+      "num_input_tokens_seen": 91817952,
+      "step": 183,
+      "train_runtime": 5208.8647,
+      "train_tokens_per_second": 17627.249
+    },
+    {
+      "epoch": 4.859060402684563,
+      "grad_norm": 1.149675965309143,
+      "learning_rate": 2.064500424599436e-07,
+      "loss": 0.2582,
+      "num_input_tokens_seen": 92328800,
+      "step": 184,
+      "train_runtime": 5234.9302,
+      "train_tokens_per_second": 17637.064
+    },
+    {
+      "epoch": 4.885906040268456,
+      "grad_norm": 1.0627394914627075,
+      "learning_rate": 1.517330670512629e-07,
+      "loss": 0.2364,
+      "num_input_tokens_seen": 92843328,
+      "step": 185,
+      "train_runtime": 5261.1199,
+      "train_tokens_per_second": 17647.066
+    },
+    {
+      "epoch": 4.912751677852349,
+      "grad_norm": 0.8081064820289612,
+      "learning_rate": 1.0540279752731253e-07,
+      "loss": 0.1979,
+      "num_input_tokens_seen": 93340480,
+      "step": 186,
+      "train_runtime": 5287.3612,
+      "train_tokens_per_second": 17653.509
+    },
+    {
+      "epoch": 4.939597315436242,
+      "grad_norm": 0.5786807537078857,
+      "learning_rate": 6.747487112931661e-08,
+      "loss": 0.6738,
+      "num_input_tokens_seen": 93856128,
+      "step": 187,
+      "train_runtime": 5314.1365,
+      "train_tokens_per_second": 17661.595
+    },
+    {
+      "epoch": 4.966442953020135,
+      "grad_norm": 0.6635648608207703,
+      "learning_rate": 3.796208916709565e-08,
+      "loss": 0.1781,
+      "num_input_tokens_seen": 94377760,
+      "step": 188,
+      "train_runtime": 5340.6451,
+      "train_tokens_per_second": 17671.603
+    },
+    {
+      "epoch": 4.993288590604027,
+      "grad_norm": 0.7493144273757935,
+      "learning_rate": 1.6874412698408836e-08,
+      "loss": 0.2207,
+      "num_input_tokens_seen": 94890496,
+      "step": 189,
+      "train_runtime": 5367.0696,
+      "train_tokens_per_second": 17680.131
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7493144273757935,
+      "learning_rate": 4.218959166932268e-09,
+      "loss": 0.3254,
+      "num_input_tokens_seen": 95021568,
+      "step": 190,
+      "train_runtime": 5372.17,
+      "train_tokens_per_second": 17687.744
+    },
+    {
+      "epoch": 5.0,
+      "num_input_tokens_seen": 95021568,
+      "step": 190,
+      "total_flos": 4.266242635254989e+18,
+      "train_loss": 0.7843870968410843,
+      "train_runtime": 5432.7802,
+      "train_samples_per_second": 2.186,
+      "train_steps_per_second": 0.035
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 190,
+  "num_input_tokens_seen": 95021568,
+  "num_train_epochs": 5,
+  "save_steps": 25,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.266242635254989e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

result_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5349bc5c9b24b21d7c574b41a997bd804e980f648221c7244a8918b1f521389e
+size 7800

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "num_input_tokens_seen": 95021568,
+    "total_flos": 4.266242635254989e+18,
+    "train_loss": 0.7843870968410843,
+    "train_runtime": 5432.7802,
+    "train_samples_per_second": 2.186,
+    "train_steps_per_second": 0.035
+}

trainer_state.json

@@ -0,0 +1,2014 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 25,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.026845637583892617,
+      "grad_norm": 8.237038612365723,
+      "learning_rate": 0.0,
+      "loss": 0.5587,
+      "num_input_tokens_seen": 496000,
+      "step": 1,
+      "train_runtime": 35.4925,
+      "train_tokens_per_second": 13974.782
+    },
+    {
+      "epoch": 0.053691275167785234,
+      "grad_norm": 8.948539733886719,
+      "learning_rate": 2.631578947368421e-06,
+      "loss": 0.6532,
+      "num_input_tokens_seen": 1007328,
+      "step": 2,
+      "train_runtime": 61.7059,
+      "train_tokens_per_second": 16324.653
+    },
+    {
+      "epoch": 0.08053691275167785,
+      "grad_norm": 5.373775482177734,
+      "learning_rate": 5.263157894736842e-06,
+      "loss": 0.4746,
+      "num_input_tokens_seen": 1520160,
+      "step": 3,
+      "train_runtime": 87.5894,
+      "train_tokens_per_second": 17355.526
+    },
+    {
+      "epoch": 0.10738255033557047,
+      "grad_norm": 8.458134651184082,
+      "learning_rate": 7.894736842105263e-06,
+      "loss": 0.5696,
+      "num_input_tokens_seen": 2031424,
+      "step": 4,
+      "train_runtime": 113.5046,
+      "train_tokens_per_second": 17897.28
+    },
+    {
+      "epoch": 0.1342281879194631,
+      "grad_norm": 7.278451442718506,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.5954,
+      "num_input_tokens_seen": 2542624,
+      "step": 5,
+      "train_runtime": 139.4544,
+      "train_tokens_per_second": 18232.649
+    },
+    {
+      "epoch": 0.1610738255033557,
+      "grad_norm": 5.226474285125732,
+      "learning_rate": 1.3157894736842106e-05,
+      "loss": 0.7396,
+      "num_input_tokens_seen": 3056000,
+      "step": 6,
+      "train_runtime": 165.2601,
+      "train_tokens_per_second": 18492.064
+    },
+    {
+      "epoch": 0.18791946308724833,
+      "grad_norm": 5.280333995819092,
+      "learning_rate": 1.5789473684210526e-05,
+      "loss": 0.7118,
+      "num_input_tokens_seen": 3574176,
+      "step": 7,
+      "train_runtime": 191.221,
+      "train_tokens_per_second": 18691.335
+    },
+    {
+      "epoch": 0.21476510067114093,
+      "grad_norm": 3.394676685333252,
+      "learning_rate": 1.8421052631578947e-05,
+      "loss": 0.5509,
+      "num_input_tokens_seen": 4076224,
+      "step": 8,
+      "train_runtime": 217.8993,
+      "train_tokens_per_second": 18706.918
+    },
+    {
+      "epoch": 0.24161073825503357,
+      "grad_norm": 2.9875869750976562,
+      "learning_rate": 2.105263157894737e-05,
+      "loss": 0.9419,
+      "num_input_tokens_seen": 4582944,
+      "step": 9,
+      "train_runtime": 243.8647,
+      "train_tokens_per_second": 18792.981
+    },
+    {
+      "epoch": 0.2684563758389262,
+      "grad_norm": 9.175309181213379,
+      "learning_rate": 2.368421052631579e-05,
+      "loss": 0.6846,
+      "num_input_tokens_seen": 5083968,
+      "step": 10,
+      "train_runtime": 270.1354,
+      "train_tokens_per_second": 18820.07
+    },
+    {
+      "epoch": 0.2953020134228188,
+      "grad_norm": 6.438014507293701,
+      "learning_rate": 2.6315789473684212e-05,
+      "loss": 0.5596,
+      "num_input_tokens_seen": 5608256,
+      "step": 11,
+      "train_runtime": 296.43,
+      "train_tokens_per_second": 18919.326
+    },
+    {
+      "epoch": 0.3221476510067114,
+      "grad_norm": 3.9598963260650635,
+      "learning_rate": 2.8947368421052634e-05,
+      "loss": 0.5313,
+      "num_input_tokens_seen": 6113696,
+      "step": 12,
+      "train_runtime": 322.4122,
+      "train_tokens_per_second": 18962.36
+    },
+    {
+      "epoch": 0.348993288590604,
+      "grad_norm": 4.5351409912109375,
+      "learning_rate": 3.157894736842105e-05,
+      "loss": 0.7476,
+      "num_input_tokens_seen": 6599840,
+      "step": 13,
+      "train_runtime": 348.2235,
+      "train_tokens_per_second": 18952.887
+    },
+    {
+      "epoch": 0.37583892617449666,
+      "grad_norm": 3.2552919387817383,
+      "learning_rate": 3.421052631578947e-05,
+      "loss": 0.6272,
+      "num_input_tokens_seen": 7116640,
+      "step": 14,
+      "train_runtime": 374.5779,
+      "train_tokens_per_second": 18999.09
+    },
+    {
+      "epoch": 0.40268456375838924,
+      "grad_norm": 3.1868414878845215,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 0.4918,
+      "num_input_tokens_seen": 7631552,
+      "step": 15,
+      "train_runtime": 400.4733,
+      "train_tokens_per_second": 19056.334
+    },
+    {
+      "epoch": 0.42953020134228187,
+      "grad_norm": 3.6740667819976807,
+      "learning_rate": 3.9473684210526316e-05,
+      "loss": 0.5583,
+      "num_input_tokens_seen": 8129664,
+      "step": 16,
+      "train_runtime": 426.6064,
+      "train_tokens_per_second": 19056.59
+    },
+    {
+      "epoch": 0.4563758389261745,
+      "grad_norm": 2.7044334411621094,
+      "learning_rate": 4.210526315789474e-05,
+      "loss": 0.6327,
+      "num_input_tokens_seen": 8648320,
+      "step": 17,
+      "train_runtime": 452.6115,
+      "train_tokens_per_second": 19107.599
+    },
+    {
+      "epoch": 0.48322147651006714,
+      "grad_norm": 2.975066661834717,
+      "learning_rate": 4.473684210526316e-05,
+      "loss": 0.5684,
+      "num_input_tokens_seen": 9166688,
+      "step": 18,
+      "train_runtime": 478.7852,
+      "train_tokens_per_second": 19145.721
+    },
+    {
+      "epoch": 0.5100671140939598,
+      "grad_norm": 4.676921367645264,
+      "learning_rate": 4.736842105263158e-05,
+      "loss": 0.6136,
+      "num_input_tokens_seen": 9689760,
+      "step": 19,
+      "train_runtime": 504.6638,
+      "train_tokens_per_second": 19200.427
+    },
+    {
+      "epoch": 0.5369127516778524,
+      "grad_norm": 3.6379780769348145,
+      "learning_rate": 5e-05,
+      "loss": 0.6268,
+      "num_input_tokens_seen": 10202784,
+      "step": 20,
+      "train_runtime": 530.9453,
+      "train_tokens_per_second": 19216.263
+    },
+    {
+      "epoch": 0.5637583892617449,
+      "grad_norm": 3.534153699874878,
+      "learning_rate": 4.999578104083307e-05,
+      "loss": 0.6514,
+      "num_input_tokens_seen": 10704736,
+      "step": 21,
+      "train_runtime": 556.8939,
+      "train_tokens_per_second": 19222.219
+    },
+    {
+      "epoch": 0.5906040268456376,
+      "grad_norm": 2.571445941925049,
+      "learning_rate": 4.998312558730159e-05,
+      "loss": 0.4698,
+      "num_input_tokens_seen": 11204608,
+      "step": 22,
+      "train_runtime": 583.084,
+      "train_tokens_per_second": 19216.113
+    },
+    {
+      "epoch": 0.6174496644295302,
+      "grad_norm": 4.351921558380127,
+      "learning_rate": 4.996203791083291e-05,
+      "loss": 0.6816,
+      "num_input_tokens_seen": 11702432,
+      "step": 23,
+      "train_runtime": 609.1711,
+      "train_tokens_per_second": 19210.418
+    },
+    {
+      "epoch": 0.6442953020134228,
+      "grad_norm": 2.935748338699341,
+      "learning_rate": 4.993252512887069e-05,
+      "loss": 0.6014,
+      "num_input_tokens_seen": 12210336,
+      "step": 24,
+      "train_runtime": 635.5688,
+      "train_tokens_per_second": 19211.666
+    },
+    {
+      "epoch": 0.6711409395973155,
+      "grad_norm": 10.634203910827637,
+      "learning_rate": 4.9894597202472696e-05,
+      "loss": 2.6525,
+      "num_input_tokens_seen": 12725728,
+      "step": 25,
+      "train_runtime": 661.4676,
+      "train_tokens_per_second": 19238.625
+    },
+    {
+      "epoch": 0.6711409395973155,
+      "eval_accuracy": 0.7960947463461368,
+      "eval_loss": 0.7538173198699951,
+      "eval_runtime": 10.0636,
+      "eval_samples_per_second": 12.421,
+      "eval_steps_per_second": 0.795,
+      "num_input_tokens_seen": 12725728,
+      "step": 25
+    },
+    {
+      "epoch": 0.697986577181208,
+      "grad_norm": 187.8636016845703,
+      "learning_rate": 4.984826693294874e-05,
+      "loss": 3.0007,
+      "num_input_tokens_seen": 13241920,
+      "step": 26,
+      "train_runtime": 762.7143,
+      "train_tokens_per_second": 17361.574
+    },
+    {
+      "epoch": 0.7248322147651006,
+      "grad_norm": 14.758489608764648,
+      "learning_rate": 4.979354995754006e-05,
+      "loss": 1.1118,
+      "num_input_tokens_seen": 13754432,
+      "step": 27,
+      "train_runtime": 788.7937,
+      "train_tokens_per_second": 17437.3
+    },
+    {
+      "epoch": 0.7516778523489933,
+      "grad_norm": 137.02674865722656,
+      "learning_rate": 4.9730464744141445e-05,
+      "loss": 5.0379,
+      "num_input_tokens_seen": 14269824,
+      "step": 28,
+      "train_runtime": 814.7738,
+      "train_tokens_per_second": 17513.848
+    },
+    {
+      "epoch": 0.7785234899328859,
+      "grad_norm": 36.753379821777344,
+      "learning_rate": 4.965903258506806e-05,
+      "loss": 1.1003,
+      "num_input_tokens_seen": 14756000,
+      "step": 29,
+      "train_runtime": 840.804,
+      "train_tokens_per_second": 17549.868
+    },
+    {
+      "epoch": 0.8053691275167785,
+      "grad_norm": 31.222320556640625,
+      "learning_rate": 4.957927758986888e-05,
+      "loss": 1.1593,
+      "num_input_tokens_seen": 15256512,
+      "step": 30,
+      "train_runtime": 867.0437,
+      "train_tokens_per_second": 17596.012
+    },
+    {
+      "epoch": 0.8322147651006712,
+      "grad_norm": 6.577232837677002,
+      "learning_rate": 4.949122667718935e-05,
+      "loss": 0.6791,
+      "num_input_tokens_seen": 15778720,
+      "step": 31,
+      "train_runtime": 892.9916,
+      "train_tokens_per_second": 17669.506
+    },
+    {
+      "epoch": 0.8590604026845637,
+      "grad_norm": 2.800877809524536,
+      "learning_rate": 4.9394909565685894e-05,
+      "loss": 0.8403,
+      "num_input_tokens_seen": 16291104,
+      "step": 32,
+      "train_runtime": 918.733,
+      "train_tokens_per_second": 17732.141
+    },
+    {
+      "epoch": 0.8859060402684564,
+      "grad_norm": 4.291952610015869,
+      "learning_rate": 4.929035876399535e-05,
+      "loss": 0.6432,
+      "num_input_tokens_seen": 16811968,
+      "step": 33,
+      "train_runtime": 944.6114,
+      "train_tokens_per_second": 17797.761
+    },
+    {
+      "epoch": 0.912751677852349,
+      "grad_norm": 2.574287176132202,
+      "learning_rate": 4.917760955976277e-05,
+      "loss": 0.5564,
+      "num_input_tokens_seen": 17327040,
+      "step": 34,
+      "train_runtime": 970.7914,
+      "train_tokens_per_second": 17848.366
+    },
+    {
+      "epoch": 0.9395973154362416,
+      "grad_norm": 57.77891159057617,
+      "learning_rate": 4.905670000773126e-05,
+      "loss": 1.8105,
+      "num_input_tokens_seen": 17823168,
+      "step": 35,
+      "train_runtime": 996.8832,
+      "train_tokens_per_second": 17878.892
+    },
+    {
+      "epoch": 0.9664429530201343,
+      "grad_norm": 6.420737266540527,
+      "learning_rate": 4.892767091689786e-05,
+      "loss": 0.9537,
+      "num_input_tokens_seen": 18343840,
+      "step": 36,
+      "train_runtime": 1023.1655,
+      "train_tokens_per_second": 17928.516
+    },
+    {
+      "epoch": 0.9932885906040269,
+      "grad_norm": 3.768702507019043,
+      "learning_rate": 4.87905658367398e-05,
+      "loss": 0.7753,
+      "num_input_tokens_seen": 18864096,
+      "step": 37,
+      "train_runtime": 1049.136,
+      "train_tokens_per_second": 17980.601
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 3.768702507019043,
+      "learning_rate": 4.864543104251587e-05,
+      "loss": 0.3534,
+      "num_input_tokens_seen": 18979328,
+      "step": 38,
+      "train_runtime": 1054.127,
+      "train_tokens_per_second": 18004.784
+    },
+    {
+      "epoch": 1.0268456375838926,
+      "grad_norm": 7.989157199859619,
+      "learning_rate": 4.849231551964771e-05,
+      "loss": 1.6501,
+      "num_input_tokens_seen": 19477504,
+      "step": 39,
+      "train_runtime": 1080.4599,
+      "train_tokens_per_second": 18027.05
+    },
+    {
+      "epoch": 1.0536912751677852,
+      "grad_norm": 3.0987582206726074,
+      "learning_rate": 4.833127094718643e-05,
+      "loss": 2.3538,
+      "num_input_tokens_seen": 19985312,
+      "step": 40,
+      "train_runtime": 1106.2719,
+      "train_tokens_per_second": 18065.462
+    },
+    {
+      "epoch": 1.0805369127516777,
+      "grad_norm": 2.594900608062744,
+      "learning_rate": 4.8162351680370044e-05,
+      "loss": 0.5087,
+      "num_input_tokens_seen": 20476800,
+      "step": 41,
+      "train_runtime": 1132.3862,
+      "train_tokens_per_second": 18082.877
+    },
+    {
+      "epoch": 1.1073825503355705,
+      "grad_norm": 3.6039469242095947,
+      "learning_rate": 4.79856147322777e-05,
+      "loss": 0.5662,
+      "num_input_tokens_seen": 20980288,
+      "step": 42,
+      "train_runtime": 1158.4548,
+      "train_tokens_per_second": 18110.58
+    },
+    {
+      "epoch": 1.1342281879194631,
+      "grad_norm": 2.1966166496276855,
+      "learning_rate": 4.7801119754586766e-05,
+      "loss": 0.4659,
+      "num_input_tokens_seen": 21492736,
+      "step": 43,
+      "train_runtime": 1184.45,
+      "train_tokens_per_second": 18145.752
+    },
+    {
+      "epoch": 1.1610738255033557,
+      "grad_norm": 2.154080629348755,
+      "learning_rate": 4.760892901743944e-05,
+      "loss": 1.9776,
+      "num_input_tokens_seen": 21997856,
+      "step": 44,
+      "train_runtime": 1210.4933,
+      "train_tokens_per_second": 18172.637
+    },
+    {
+      "epoch": 1.1879194630872483,
+      "grad_norm": 68.91860961914062,
+      "learning_rate": 4.7409107388425504e-05,
+      "loss": 3.0839,
+      "num_input_tokens_seen": 22509088,
+      "step": 45,
+      "train_runtime": 1236.7017,
+      "train_tokens_per_second": 18200.903
+    },
+    {
+      "epoch": 1.2147651006711409,
+      "grad_norm": 24.943208694458008,
+      "learning_rate": 4.7201722310688445e-05,
+      "loss": 0.6674,
+      "num_input_tokens_seen": 23024384,
+      "step": 46,
+      "train_runtime": 1263.3678,
+      "train_tokens_per_second": 18224.609
+    },
+    {
+      "epoch": 1.2416107382550337,
+      "grad_norm": 5.5658488273620605,
+      "learning_rate": 4.698684378016222e-05,
+      "loss": 1.0302,
+      "num_input_tokens_seen": 23536384,
+      "step": 47,
+      "train_runtime": 1289.0088,
+      "train_tokens_per_second": 18259.289
+    },
+    {
+      "epoch": 1.2684563758389262,
+      "grad_norm": 34.10597229003906,
+      "learning_rate": 4.676454432194656e-05,
+      "loss": 1.3633,
+      "num_input_tokens_seen": 24046432,
+      "step": 48,
+      "train_runtime": 1315.3663,
+      "train_tokens_per_second": 18281.167
+    },
+    {
+      "epoch": 1.2953020134228188,
+      "grad_norm": 3.001063585281372,
+      "learning_rate": 4.6534898965828405e-05,
+      "loss": 0.5103,
+      "num_input_tokens_seen": 24549600,
+      "step": 49,
+      "train_runtime": 1341.5667,
+      "train_tokens_per_second": 18299.203
+    },
+    {
+      "epoch": 1.3221476510067114,
+      "grad_norm": 2.9493746757507324,
+      "learning_rate": 4.629798522095818e-05,
+      "loss": 0.6796,
+      "num_input_tokens_seen": 25037312,
+      "step": 50,
+      "train_runtime": 1367.4664,
+      "train_tokens_per_second": 18309.27
+    },
+    {
+      "epoch": 1.3221476510067114,
+      "eval_accuracy": 0.8520892900119574,
+      "eval_loss": 0.5981780886650085,
+      "eval_runtime": 10.0754,
+      "eval_samples_per_second": 12.406,
+      "eval_steps_per_second": 0.794,
+      "num_input_tokens_seen": 25037312,
+      "step": 50
+    },
+    {
+      "epoch": 1.348993288590604,
+      "grad_norm": 1.894378423690796,
+      "learning_rate": 4.6053883049689145e-05,
+      "loss": 0.7632,
+      "num_input_tokens_seen": 25548352,
+      "step": 51,
+      "train_runtime": 1462.8406,
+      "train_tokens_per_second": 17464.891
+    },
+    {
+      "epoch": 1.3758389261744965,
+      "grad_norm": 3.0130774974823,
+      "learning_rate": 4.580267484058876e-05,
+      "loss": 0.5686,
+      "num_input_tokens_seen": 26063136,
+      "step": 52,
+      "train_runtime": 1489.1172,
+      "train_tokens_per_second": 17502.407
+    },
+    {
+      "epoch": 1.4026845637583891,
+      "grad_norm": 2.3910205364227295,
+      "learning_rate": 4.554444538063113e-05,
+      "loss": 0.5809,
+      "num_input_tokens_seen": 26574336,
+      "step": 53,
+      "train_runtime": 1515.4499,
+      "train_tokens_per_second": 17535.608
+    },
+    {
+      "epoch": 1.429530201342282,
+      "grad_norm": 2.2223563194274902,
+      "learning_rate": 4.5279281826580056e-05,
+      "loss": 0.4817,
+      "num_input_tokens_seen": 27078496,
+      "step": 54,
+      "train_runtime": 1541.8316,
+      "train_tokens_per_second": 17562.551
+    },
+    {
+      "epoch": 1.4563758389261745,
+      "grad_norm": 1.928465485572815,
+      "learning_rate": 4.5007273675572104e-05,
+      "loss": 0.4111,
+      "num_input_tokens_seen": 27599232,
+      "step": 55,
+      "train_runtime": 1568.1389,
+      "train_tokens_per_second": 17599.992
+    },
+    {
+      "epoch": 1.483221476510067,
+      "grad_norm": 2.8375704288482666,
+      "learning_rate": 4.4728512734909844e-05,
+      "loss": 0.5102,
+      "num_input_tokens_seen": 28099968,
+      "step": 56,
+      "train_runtime": 1594.4215,
+      "train_tokens_per_second": 17623.927
+    },
+    {
+      "epoch": 1.5100671140939599,
+      "grad_norm": 2.123541831970215,
+      "learning_rate": 4.444309309107535e-05,
+      "loss": 0.8591,
+      "num_input_tokens_seen": 28604864,
+      "step": 57,
+      "train_runtime": 1620.7434,
+      "train_tokens_per_second": 17649.224
+    },
+    {
+      "epoch": 1.5369127516778525,
+      "grad_norm": 7.7735066413879395,
+      "learning_rate": 4.415111107797445e-05,
+      "loss": 0.6124,
+      "num_input_tokens_seen": 29112480,
+      "step": 58,
+      "train_runtime": 1646.8857,
+      "train_tokens_per_second": 17677.293
+    },
+    {
+      "epoch": 1.563758389261745,
+      "grad_norm": 51.44823455810547,
+      "learning_rate": 4.385266524442241e-05,
+      "loss": 0.8082,
+      "num_input_tokens_seen": 29610208,
+      "step": 59,
+      "train_runtime": 1672.6373,
+      "train_tokens_per_second": 17702.707
+    },
+    {
+      "epoch": 1.5906040268456376,
+      "grad_norm": 2.954522132873535,
+      "learning_rate": 4.3547856320882044e-05,
+      "loss": 0.5294,
+      "num_input_tokens_seen": 30121920,
+      "step": 60,
+      "train_runtime": 1698.4583,
+      "train_tokens_per_second": 17734.86
+    },
+    {
+      "epoch": 1.6174496644295302,
+      "grad_norm": 2.5693578720092773,
+      "learning_rate": 4.3236787185465525e-05,
+      "loss": 0.5192,
+      "num_input_tokens_seen": 30632928,
+      "step": 61,
+      "train_runtime": 1724.3917,
+      "train_tokens_per_second": 17764.484
+    },
+    {
+      "epoch": 1.6442953020134228,
+      "grad_norm": 4.527597427368164,
+      "learning_rate": 4.2919562829211283e-05,
+      "loss": 0.554,
+      "num_input_tokens_seen": 31139872,
+      "step": 62,
+      "train_runtime": 1750.568,
+      "train_tokens_per_second": 17788.439
+    },
+    {
+      "epoch": 1.6711409395973154,
+      "grad_norm": 1.5587811470031738,
+      "learning_rate": 4.259629032064779e-05,
+      "loss": 0.4231,
+      "num_input_tokens_seen": 31645376,
+      "step": 63,
+      "train_runtime": 1776.4231,
+      "train_tokens_per_second": 17814.098
+    },
+    {
+      "epoch": 1.697986577181208,
+      "grad_norm": 2.3483505249023438,
+      "learning_rate": 4.226707876965611e-05,
+      "loss": 1.475,
+      "num_input_tokens_seen": 32167168,
+      "step": 64,
+      "train_runtime": 1802.9115,
+      "train_tokens_per_second": 17841.79
+    },
+    {
+      "epoch": 1.7248322147651005,
+      "grad_norm": 104.57856750488281,
+      "learning_rate": 4.193203929064353e-05,
+      "loss": 1.6005,
+      "num_input_tokens_seen": 32691456,
+      "step": 65,
+      "train_runtime": 1829.164,
+      "train_tokens_per_second": 17872.348
+    },
+    {
+      "epoch": 1.7516778523489933,
+      "grad_norm": 2.754912853240967,
+      "learning_rate": 4.159128496504053e-05,
+      "loss": 1.1018,
+      "num_input_tokens_seen": 33188672,
+      "step": 66,
+      "train_runtime": 1855.5298,
+      "train_tokens_per_second": 17886.359
+    },
+    {
+      "epoch": 1.778523489932886,
+      "grad_norm": 2.767381429672241,
+      "learning_rate": 4.1244930803134e-05,
+      "loss": 1.2071,
+      "num_input_tokens_seen": 33690880,
+      "step": 67,
+      "train_runtime": 1881.8777,
+      "train_tokens_per_second": 17902.8
+    },
+    {
+      "epoch": 1.8053691275167785,
+      "grad_norm": 2.260093927383423,
+      "learning_rate": 4.089309370524921e-05,
+      "loss": 0.6214,
+      "num_input_tokens_seen": 34196736,
+      "step": 68,
+      "train_runtime": 1907.9434,
+      "train_tokens_per_second": 17923.349
+    },
+    {
+      "epoch": 1.8322147651006713,
+      "grad_norm": 2.873724937438965,
+      "learning_rate": 4.053589242229412e-05,
+      "loss": 0.5573,
+      "num_input_tokens_seen": 34720768,
+      "step": 69,
+      "train_runtime": 1934.0333,
+      "train_tokens_per_second": 17952.518
+    },
+    {
+      "epoch": 1.8590604026845639,
+      "grad_norm": 2.169739007949829,
+      "learning_rate": 4.0173447515678916e-05,
+      "loss": 0.436,
+      "num_input_tokens_seen": 35230112,
+      "step": 70,
+      "train_runtime": 1959.9087,
+      "train_tokens_per_second": 17975.384
+    },
+    {
+      "epoch": 1.8859060402684564,
+      "grad_norm": 2.2907986640930176,
+      "learning_rate": 3.9805881316624506e-05,
+      "loss": 0.4724,
+      "num_input_tokens_seen": 35752064,
+      "step": 71,
+      "train_runtime": 1986.0719,
+      "train_tokens_per_second": 18001.395
+    },
+    {
+      "epoch": 1.912751677852349,
+      "grad_norm": 3.054774284362793,
+      "learning_rate": 3.9433317884873664e-05,
+      "loss": 0.4414,
+      "num_input_tokens_seen": 36274496,
+      "step": 72,
+      "train_runtime": 2012.7145,
+      "train_tokens_per_second": 18022.673
+    },
+    {
+      "epoch": 1.9395973154362416,
+      "grad_norm": 1.7803468704223633,
+      "learning_rate": 3.905588296681886e-05,
+      "loss": 0.3534,
+      "num_input_tokens_seen": 36777152,
+      "step": 73,
+      "train_runtime": 2039.1555,
+      "train_tokens_per_second": 18035.482
+    },
+    {
+      "epoch": 1.9664429530201342,
+      "grad_norm": 111.7915267944336,
+      "learning_rate": 3.867370395306068e-05,
+      "loss": 1.1611,
+      "num_input_tokens_seen": 37298112,
+      "step": 74,
+      "train_runtime": 2065.3775,
+      "train_tokens_per_second": 18058.738
+    },
+    {
+      "epoch": 1.9932885906040267,
+      "grad_norm": 2.8219268321990967,
+      "learning_rate": 3.82869098354114e-05,
+      "loss": 0.6056,
+      "num_input_tokens_seen": 37817120,
+      "step": 75,
+      "train_runtime": 2092.2846,
+      "train_tokens_per_second": 18074.558
+    },
+    {
+      "epoch": 1.9932885906040267,
+      "eval_accuracy": 0.8556461964204233,
+      "eval_loss": 0.5577213764190674,
+      "eval_runtime": 10.0688,
+      "eval_samples_per_second": 12.415,
+      "eval_steps_per_second": 0.795,
+      "num_input_tokens_seen": 37817120,
+      "step": 75
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.8219268321990967,
+      "learning_rate": 3.7895631163358105e-05,
+      "loss": 0.4723,
+      "num_input_tokens_seen": 37948192,
+      "step": 76,
+      "train_runtime": 2167.0471,
+      "train_tokens_per_second": 17511.475
+    },
+    {
+      "epoch": 2.0268456375838926,
+      "grad_norm": 2.0256853103637695,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.4424,
+      "num_input_tokens_seen": 38467968,
+      "step": 77,
+      "train_runtime": 2193.3126,
+      "train_tokens_per_second": 17538.753
+    },
+    {
+      "epoch": 2.053691275167785,
+      "grad_norm": 1.5646251440048218,
+      "learning_rate": 3.7100149877474974e-05,
+      "loss": 1.4039,
+      "num_input_tokens_seen": 38986496,
+      "step": 78,
+      "train_runtime": 2219.7314,
+      "train_tokens_per_second": 17563.609
+    },
+    {
+      "epoch": 2.0805369127516777,
+      "grad_norm": 2.312363386154175,
+      "learning_rate": 3.66962157518902e-05,
+      "loss": 0.3641,
+      "num_input_tokens_seen": 39507008,
+      "step": 79,
+      "train_runtime": 2246.4116,
+      "train_tokens_per_second": 17586.718
+    },
+    {
+      "epoch": 2.1073825503355703,
+      "grad_norm": 1.7589001655578613,
+      "learning_rate": 3.628833395777224e-05,
+      "loss": 0.3558,
+      "num_input_tokens_seen": 40012992,
+      "step": 80,
+      "train_runtime": 2272.6092,
+      "train_tokens_per_second": 17606.632
+    },
+    {
+      "epoch": 2.134228187919463,
+      "grad_norm": 1.479117751121521,
+      "learning_rate": 3.587664216205183e-05,
+      "loss": 0.4281,
+      "num_input_tokens_seen": 40536608,
+      "step": 81,
+      "train_runtime": 2298.9106,
+      "train_tokens_per_second": 17632.964
+    },
+    {
+      "epoch": 2.1610738255033555,
+      "grad_norm": 1.605286717414856,
+      "learning_rate": 3.546127931759903e-05,
+      "loss": 0.3695,
+      "num_input_tokens_seen": 41052192,
+      "step": 82,
+      "train_runtime": 2324.8418,
+      "train_tokens_per_second": 17658.058
+    },
+    {
+      "epoch": 2.1879194630872485,
+      "grad_norm": 19.489822387695312,
+      "learning_rate": 3.504238561632424e-05,
+      "loss": 0.967,
+      "num_input_tokens_seen": 41559584,
+      "step": 83,
+      "train_runtime": 2351.0304,
+      "train_tokens_per_second": 17677.179
+    },
+    {
+      "epoch": 2.214765100671141,
+      "grad_norm": 3.9726529121398926,
+      "learning_rate": 3.4620102441861143e-05,
+      "loss": 0.4472,
+      "num_input_tokens_seen": 42060128,
+      "step": 84,
+      "train_runtime": 2377.0613,
+      "train_tokens_per_second": 17694.171
+    },
+    {
+      "epoch": 2.2416107382550337,
+      "grad_norm": 3.072654962539673,
+      "learning_rate": 3.4194572321847336e-05,
+      "loss": 1.2018,
+      "num_input_tokens_seen": 42570080,
+      "step": 85,
+      "train_runtime": 2403.2948,
+      "train_tokens_per_second": 17713.216
+    },
+    {
+      "epoch": 2.2684563758389262,
+      "grad_norm": 1.8841156959533691,
+      "learning_rate": 3.376593887981887e-05,
+      "loss": 1.1866,
+      "num_input_tokens_seen": 43068832,
+      "step": 86,
+      "train_runtime": 2429.5267,
+      "train_tokens_per_second": 17727.252
+    },
+    {
+      "epoch": 2.295302013422819,
+      "grad_norm": 58.6556396484375,
+      "learning_rate": 3.333434678673489e-05,
+      "loss": 1.9914,
+      "num_input_tokens_seen": 43582016,
+      "step": 87,
+      "train_runtime": 2455.318,
+      "train_tokens_per_second": 17750.049
+    },
+    {
+      "epoch": 2.3221476510067114,
+      "grad_norm": 3.74403715133667,
+      "learning_rate": 3.289994171214882e-05,
+      "loss": 0.6787,
+      "num_input_tokens_seen": 44089344,
+      "step": 88,
+      "train_runtime": 2481.5665,
+      "train_tokens_per_second": 17766.739
+    },
+    {
+      "epoch": 2.348993288590604,
+      "grad_norm": 3.9329142570495605,
+      "learning_rate": 3.246287027504237e-05,
+      "loss": 0.4743,
+      "num_input_tokens_seen": 44589408,
+      "step": 89,
+      "train_runtime": 2507.2669,
+      "train_tokens_per_second": 17784.069
+    },
+    {
+      "epoch": 2.3758389261744965,
+      "grad_norm": 1.8977859020233154,
+      "learning_rate": 3.202327999433924e-05,
+      "loss": 0.4042,
+      "num_input_tokens_seen": 45105888,
+      "step": 90,
+      "train_runtime": 2533.214,
+      "train_tokens_per_second": 17805.795
+    },
+    {
+      "epoch": 2.402684563758389,
+      "grad_norm": 1.4344909191131592,
+      "learning_rate": 3.158131923911498e-05,
+      "loss": 0.4187,
+      "num_input_tokens_seen": 45623968,
+      "step": 91,
+      "train_runtime": 2559.4997,
+      "train_tokens_per_second": 17825.346
+    },
+    {
+      "epoch": 2.4295302013422817,
+      "grad_norm": 2.5110371112823486,
+      "learning_rate": 3.1137137178519985e-05,
+      "loss": 3.5044,
+      "num_input_tokens_seen": 46136544,
+      "step": 92,
+      "train_runtime": 2585.6658,
+      "train_tokens_per_second": 17843.197
+    },
+    {
+      "epoch": 2.4563758389261743,
+      "grad_norm": 1.7893449068069458,
+      "learning_rate": 3.069088373143234e-05,
+      "loss": 0.3795,
+      "num_input_tokens_seen": 46651040,
+      "step": 93,
+      "train_runtime": 2611.8151,
+      "train_tokens_per_second": 17861.54
+    },
+    {
+      "epoch": 2.4832214765100673,
+      "grad_norm": 176.00808715820312,
+      "learning_rate": 3.0242709515857758e-05,
+      "loss": 2.3593,
+      "num_input_tokens_seen": 47163616,
+      "step": 94,
+      "train_runtime": 2638.1455,
+      "train_tokens_per_second": 17877.565
+    },
+    {
+      "epoch": 2.51006711409396,
+      "grad_norm": 884.7402954101562,
+      "learning_rate": 2.9792765798093465e-05,
+      "loss": 1.9348,
+      "num_input_tokens_seen": 47674976,
+      "step": 95,
+      "train_runtime": 2663.9769,
+      "train_tokens_per_second": 17896.167
+    },
+    {
+      "epoch": 2.5369127516778525,
+      "grad_norm": 153.87149047851562,
+      "learning_rate": 2.9341204441673266e-05,
+      "loss": 5.973,
+      "num_input_tokens_seen": 48185504,
+      "step": 96,
+      "train_runtime": 2690.5787,
+      "train_tokens_per_second": 17908.974
+    },
+    {
+      "epoch": 2.563758389261745,
+      "grad_norm": 3.753488302230835,
+      "learning_rate": 2.8888177856111083e-05,
+      "loss": 0.5242,
+      "num_input_tokens_seen": 48689920,
+      "step": 97,
+      "train_runtime": 2716.6014,
+      "train_tokens_per_second": 17923.1
+    },
+    {
+      "epoch": 2.5906040268456376,
+      "grad_norm": 4.23100471496582,
+      "learning_rate": 2.8433838945460205e-05,
+      "loss": 0.6778,
+      "num_input_tokens_seen": 49207680,
+      "step": 98,
+      "train_runtime": 2742.4391,
+      "train_tokens_per_second": 17943.034
+    },
+    {
+      "epoch": 2.61744966442953,
+      "grad_norm": 2.3302712440490723,
+      "learning_rate": 2.797834105670559e-05,
+      "loss": 0.4994,
+      "num_input_tokens_seen": 49701344,
+      "step": 99,
+      "train_runtime": 2768.5965,
+      "train_tokens_per_second": 17951.819
+    },
+    {
+      "epoch": 2.6442953020134228,
+      "grad_norm": 1.5278692245483398,
+      "learning_rate": 2.752183792800671e-05,
+      "loss": 0.4019,
+      "num_input_tokens_seen": 50216704,
+      "step": 100,
+      "train_runtime": 2794.7132,
+      "train_tokens_per_second": 17968.464
+    },
+    {
+      "epoch": 2.6442953020134228,
+      "eval_accuracy": 0.868777035442255,
+      "eval_loss": 0.5278336405754089,
+      "eval_runtime": 10.0705,
+      "eval_samples_per_second": 12.412,
+      "eval_steps_per_second": 0.794,
+      "num_input_tokens_seen": 50216704,
+      "step": 100
+    },
+    {
+      "epoch": 2.6711409395973154,
+      "grad_norm": 1.9811456203460693,
+      "learning_rate": 2.7064483636808313e-05,
+      "loss": 1.1764,
+      "num_input_tokens_seen": 50727392,
+      "step": 101,
+      "train_runtime": 2890.7352,
+      "train_tokens_per_second": 17548.267
+    },
+    {
+      "epoch": 2.697986577181208,
+      "grad_norm": 22.440269470214844,
+      "learning_rate": 2.6606432547836757e-05,
+      "loss": 0.8064,
+      "num_input_tokens_seen": 51242496,
+      "step": 102,
+      "train_runtime": 2917.1561,
+      "train_tokens_per_second": 17565.908
+    },
+    {
+      "epoch": 2.7248322147651005,
+      "grad_norm": 1.9680432081222534,
+      "learning_rate": 2.6147839260999295e-05,
+      "loss": 6.5702,
+      "num_input_tokens_seen": 51725728,
+      "step": 103,
+      "train_runtime": 2943.046,
+      "train_tokens_per_second": 17575.575
+    },
+    {
+      "epoch": 2.751677852348993,
+      "grad_norm": 74.59529113769531,
+      "learning_rate": 2.5688858559204053e-05,
+      "loss": 0.3993,
+      "num_input_tokens_seen": 52244192,
+      "step": 104,
+      "train_runtime": 2969.7783,
+      "train_tokens_per_second": 17591.95
+    },
+    {
+      "epoch": 2.778523489932886,
+      "grad_norm": 1.6544747352600098,
+      "learning_rate": 2.5229645356118163e-05,
+      "loss": 0.4268,
+      "num_input_tokens_seen": 52757856,
+      "step": 105,
+      "train_runtime": 2996.1745,
+      "train_tokens_per_second": 17608.405
+    },
+    {
+      "epoch": 2.8053691275167782,
+      "grad_norm": 1.4038758277893066,
+      "learning_rate": 2.4770354643881843e-05,
+      "loss": 0.358,
+      "num_input_tokens_seen": 53267776,
+      "step": 106,
+      "train_runtime": 3022.3818,
+      "train_tokens_per_second": 17624.436
+    },
+    {
+      "epoch": 2.8322147651006713,
+      "grad_norm": 1.7984601259231567,
+      "learning_rate": 2.4311141440795953e-05,
+      "loss": 1.1325,
+      "num_input_tokens_seen": 53774464,
+      "step": 107,
+      "train_runtime": 3048.432,
+      "train_tokens_per_second": 17640.041
+    },
+    {
+      "epoch": 2.859060402684564,
+      "grad_norm": 40.03776931762695,
+      "learning_rate": 2.3852160739000707e-05,
+      "loss": 3.3749,
+      "num_input_tokens_seen": 54287936,
+      "step": 108,
+      "train_runtime": 3074.3539,
+      "train_tokens_per_second": 17658.324
+    },
+    {
+      "epoch": 2.8859060402684564,
+      "grad_norm": 2.0243523120880127,
+      "learning_rate": 2.3393567452163252e-05,
+      "loss": 0.3589,
+      "num_input_tokens_seen": 54776736,
+      "step": 109,
+      "train_runtime": 3100.4425,
+      "train_tokens_per_second": 17667.393
+    },
+    {
+      "epoch": 2.912751677852349,
+      "grad_norm": 1.841299057006836,
+      "learning_rate": 2.2935516363191693e-05,
+      "loss": 0.4193,
+      "num_input_tokens_seen": 55271424,
+      "step": 110,
+      "train_runtime": 3126.5372,
+      "train_tokens_per_second": 17678.16
+    },
+    {
+      "epoch": 2.9395973154362416,
+      "grad_norm": 1.390844702720642,
+      "learning_rate": 2.2478162071993298e-05,
+      "loss": 0.3847,
+      "num_input_tokens_seen": 55769824,
+      "step": 111,
+      "train_runtime": 3152.3296,
+      "train_tokens_per_second": 17691.622
+    },
+    {
+      "epoch": 2.966442953020134,
+      "grad_norm": 1.4262468814849854,
+      "learning_rate": 2.202165894329441e-05,
+      "loss": 0.3429,
+      "num_input_tokens_seen": 56289728,
+      "step": 112,
+      "train_runtime": 3178.39,
+      "train_tokens_per_second": 17710.139
+    },
+    {
+      "epoch": 2.9932885906040267,
+      "grad_norm": 1.5515210628509521,
+      "learning_rate": 2.1566161054539798e-05,
+      "loss": 0.4288,
+      "num_input_tokens_seen": 56806016,
+      "step": 113,
+      "train_runtime": 3204.5586,
+      "train_tokens_per_second": 17726.627
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.5515210628509521,
+      "learning_rate": 2.111182214388893e-05,
+      "loss": 0.389,
+      "num_input_tokens_seen": 56937088,
+      "step": 114,
+      "train_runtime": 3209.7058,
+      "train_tokens_per_second": 17739.036
+    },
+    {
+      "epoch": 3.0268456375838926,
+      "grad_norm": 1.1605178117752075,
+      "learning_rate": 2.0658795558326743e-05,
+      "loss": 3.2378,
+      "num_input_tokens_seen": 57437376,
+      "step": 115,
+      "train_runtime": 3236.2481,
+      "train_tokens_per_second": 17748.137
+    },
+    {
+      "epoch": 3.053691275167785,
+      "grad_norm": 45.809234619140625,
+      "learning_rate": 2.0207234201906547e-05,
+      "loss": 0.7378,
+      "num_input_tokens_seen": 57926240,
+      "step": 116,
+      "train_runtime": 3262.3674,
+      "train_tokens_per_second": 17755.891
+    },
+    {
+      "epoch": 3.0805369127516777,
+      "grad_norm": 44.064762115478516,
+      "learning_rate": 1.9757290484142244e-05,
+      "loss": 0.3071,
+      "num_input_tokens_seen": 58445376,
+      "step": 117,
+      "train_runtime": 3288.5966,
+      "train_tokens_per_second": 17772.133
+    },
+    {
+      "epoch": 3.1073825503355703,
+      "grad_norm": 2.5294344425201416,
+      "learning_rate": 1.9309116268567674e-05,
+      "loss": 0.5286,
+      "num_input_tokens_seen": 58940512,
+      "step": 118,
+      "train_runtime": 3314.2764,
+      "train_tokens_per_second": 17783.825
+    },
+    {
+      "epoch": 3.134228187919463,
+      "grad_norm": 2.2777907848358154,
+      "learning_rate": 1.8862862821480025e-05,
+      "loss": 1.2341,
+      "num_input_tokens_seen": 59447232,
+      "step": 119,
+      "train_runtime": 3340.3057,
+      "train_tokens_per_second": 17796.943
+    },
+    {
+      "epoch": 3.1610738255033555,
+      "grad_norm": 1.2976305484771729,
+      "learning_rate": 1.8418680760885027e-05,
+      "loss": 0.2624,
+      "num_input_tokens_seen": 59962592,
+      "step": 120,
+      "train_runtime": 3366.4974,
+      "train_tokens_per_second": 17811.567
+    },
+    {
+      "epoch": 3.1879194630872485,
+      "grad_norm": 1.4661471843719482,
+      "learning_rate": 1.797672000566077e-05,
+      "loss": 0.3812,
+      "num_input_tokens_seen": 60473632,
+      "step": 121,
+      "train_runtime": 3392.1332,
+      "train_tokens_per_second": 17827.611
+    },
+    {
+      "epoch": 3.214765100671141,
+      "grad_norm": 1.7925889492034912,
+      "learning_rate": 1.7537129724957642e-05,
+      "loss": 0.3298,
+      "num_input_tokens_seen": 60984768,
+      "step": 122,
+      "train_runtime": 3418.1221,
+      "train_tokens_per_second": 17841.6
+    },
+    {
+      "epoch": 3.2416107382550337,
+      "grad_norm": 1.1223887205123901,
+      "learning_rate": 1.710005828785119e-05,
+      "loss": 0.316,
+      "num_input_tokens_seen": 61507680,
+      "step": 123,
+      "train_runtime": 3444.5252,
+      "train_tokens_per_second": 17856.65
+    },
+    {
+      "epoch": 3.2684563758389262,
+      "grad_norm": 1.381909966468811,
+      "learning_rate": 1.666565321326512e-05,
+      "loss": 0.8116,
+      "num_input_tokens_seen": 62023520,
+      "step": 124,
+      "train_runtime": 3470.5519,
+      "train_tokens_per_second": 17871.371
+    },
+    {
+      "epoch": 3.295302013422819,
+      "grad_norm": 40.172916412353516,
+      "learning_rate": 1.6234061120181142e-05,
+      "loss": 0.3133,
+      "num_input_tokens_seen": 62538624,
+      "step": 125,
+      "train_runtime": 3496.8506,
+      "train_tokens_per_second": 17884.271
+    },
+    {
+      "epoch": 3.295302013422819,
+      "eval_accuracy": 0.8777751968802215,
+      "eval_loss": 0.4720778465270996,
+      "eval_runtime": 10.0701,
+      "eval_samples_per_second": 12.413,
+      "eval_steps_per_second": 0.794,
+      "num_input_tokens_seen": 62538624,
+      "step": 125
+    },
+    {
+      "epoch": 3.3221476510067114,
+      "grad_norm": 1.1484184265136719,
+      "learning_rate": 1.5805427678152677e-05,
+      "loss": 0.8765,
+      "num_input_tokens_seen": 63047392,
+      "step": 126,
+      "train_runtime": 3593.249,
+      "train_tokens_per_second": 17546.068
+    },
+    {
+      "epoch": 3.348993288590604,
+      "grad_norm": 0.9760416746139526,
+      "learning_rate": 1.5379897558138862e-05,
+      "loss": 0.2751,
+      "num_input_tokens_seen": 63547936,
+      "step": 127,
+      "train_runtime": 3620.1181,
+      "train_tokens_per_second": 17554.106
+    },
+    {
+      "epoch": 3.3758389261744965,
+      "grad_norm": 41.91545867919922,
+      "learning_rate": 1.495761438367577e-05,
+      "loss": 1.2805,
+      "num_input_tokens_seen": 64051232,
+      "step": 128,
+      "train_runtime": 3646.4211,
+      "train_tokens_per_second": 17565.506
+    },
+    {
+      "epoch": 3.402684563758389,
+      "grad_norm": 1.1068097352981567,
+      "learning_rate": 1.4538720682400969e-05,
+      "loss": 0.2843,
+      "num_input_tokens_seen": 64546048,
+      "step": 129,
+      "train_runtime": 3673.1876,
+      "train_tokens_per_second": 17572.217
+    },
+    {
+      "epoch": 3.4295302013422817,
+      "grad_norm": 1.142320156097412,
+      "learning_rate": 1.4123357837948175e-05,
+      "loss": 0.3334,
+      "num_input_tokens_seen": 65047392,
+      "step": 130,
+      "train_runtime": 3699.4486,
+      "train_tokens_per_second": 17582.997
+    },
+    {
+      "epoch": 3.4563758389261743,
+      "grad_norm": 46.183719635009766,
+      "learning_rate": 1.3711666042227772e-05,
+      "loss": 1.2229,
+      "num_input_tokens_seen": 65565600,
+      "step": 131,
+      "train_runtime": 3725.8008,
+      "train_tokens_per_second": 17597.72
+    },
+    {
+      "epoch": 3.4832214765100673,
+      "grad_norm": 0.8785358667373657,
+      "learning_rate": 1.330378424810981e-05,
+      "loss": 0.4387,
+      "num_input_tokens_seen": 66075392,
+      "step": 132,
+      "train_runtime": 3752.0888,
+      "train_tokens_per_second": 17610.295
+    },
+    {
+      "epoch": 3.51006711409396,
+      "grad_norm": 1.0320525169372559,
+      "learning_rate": 1.2899850122525037e-05,
+      "loss": 0.2675,
+      "num_input_tokens_seen": 66591552,
+      "step": 133,
+      "train_runtime": 3777.8237,
+      "train_tokens_per_second": 17626.961
+    },
+    {
+      "epoch": 3.5369127516778525,
+      "grad_norm": 1.3758822679519653,
+      "learning_rate": 1.2500000000000006e-05,
+      "loss": 0.2816,
+      "num_input_tokens_seen": 67106688,
+      "step": 134,
+      "train_runtime": 3803.8322,
+      "train_tokens_per_second": 17641.863
+    },
+    {
+      "epoch": 3.563758389261745,
+      "grad_norm": 1.0216566324234009,
+      "learning_rate": 1.2104368836641908e-05,
+      "loss": 0.2957,
+      "num_input_tokens_seen": 67618688,
+      "step": 135,
+      "train_runtime": 3829.8553,
+      "train_tokens_per_second": 17655.677
+    },
+    {
+      "epoch": 3.5906040268456376,
+      "grad_norm": 91.42462921142578,
+      "learning_rate": 1.1713090164588607e-05,
+      "loss": 2.8051,
+      "num_input_tokens_seen": 68129952,
+      "step": 136,
+      "train_runtime": 3856.2115,
+      "train_tokens_per_second": 17667.587
+    },
+    {
+      "epoch": 3.61744966442953,
+      "grad_norm": 0.8375159502029419,
+      "learning_rate": 1.1326296046939333e-05,
+      "loss": 0.2361,
+      "num_input_tokens_seen": 68642048,
+      "step": 137,
+      "train_runtime": 3882.2608,
+      "train_tokens_per_second": 17680.947
+    },
+    {
+      "epoch": 3.6442953020134228,
+      "grad_norm": 1.1549395322799683,
+      "learning_rate": 1.0944117033181151e-05,
+      "loss": 0.4118,
+      "num_input_tokens_seen": 69161312,
+      "step": 138,
+      "train_runtime": 3908.295,
+      "train_tokens_per_second": 17696.032
+    },
+    {
+      "epoch": 3.6711409395973154,
+      "grad_norm": 1.7038424015045166,
+      "learning_rate": 1.0566682115126344e-05,
+      "loss": 0.3521,
+      "num_input_tokens_seen": 69669024,
+      "step": 139,
+      "train_runtime": 3934.6575,
+      "train_tokens_per_second": 17706.503
+    },
+    {
+      "epoch": 3.697986577181208,
+      "grad_norm": 0.8558218479156494,
+      "learning_rate": 1.0194118683375503e-05,
+      "loss": 0.3169,
+      "num_input_tokens_seen": 70169152,
+      "step": 140,
+      "train_runtime": 3960.2013,
+      "train_tokens_per_second": 17718.582
+    },
+    {
+      "epoch": 3.7248322147651005,
+      "grad_norm": 1.4320504665374756,
+      "learning_rate": 9.826552484321087e-06,
+      "loss": 0.2943,
+      "num_input_tokens_seen": 70691424,
+      "step": 141,
+      "train_runtime": 3986.6426,
+      "train_tokens_per_second": 17732.069
+    },
+    {
+      "epoch": 3.751677852348993,
+      "grad_norm": 1.188143253326416,
+      "learning_rate": 9.464107577705886e-06,
+      "loss": 0.4079,
+      "num_input_tokens_seen": 71198976,
+      "step": 142,
+      "train_runtime": 4012.8509,
+      "train_tokens_per_second": 17742.742
+    },
+    {
+      "epoch": 3.778523489932886,
+      "grad_norm": 1.1367772817611694,
+      "learning_rate": 9.106906294750805e-06,
+      "loss": 0.2099,
+      "num_input_tokens_seen": 71713312,
+      "step": 143,
+      "train_runtime": 4039.1047,
+      "train_tokens_per_second": 17754.754
+    },
+    {
+      "epoch": 3.8053691275167782,
+      "grad_norm": 0.9287247657775879,
+      "learning_rate": 8.755069196866014e-06,
+      "loss": 0.3496,
+      "num_input_tokens_seen": 72232064,
+      "step": 144,
+      "train_runtime": 4065.3712,
+      "train_tokens_per_second": 17767.643
+    },
+    {
+      "epoch": 3.8322147651006713,
+      "grad_norm": 0.8527315855026245,
+      "learning_rate": 8.40871503495947e-06,
+      "loss": 0.2696,
+      "num_input_tokens_seen": 72743648,
+      "step": 145,
+      "train_runtime": 4091.4699,
+      "train_tokens_per_second": 17779.343
+    },
+    {
+      "epoch": 3.859060402684564,
+      "grad_norm": 1.0350202322006226,
+      "learning_rate": 8.067960709356478e-06,
+      "loss": 0.2302,
+      "num_input_tokens_seen": 73266496,
+      "step": 146,
+      "train_runtime": 4117.9334,
+      "train_tokens_per_second": 17792.055
+    },
+    {
+      "epoch": 3.8859060402684564,
+      "grad_norm": 0.8234533667564392,
+      "learning_rate": 7.732921230343892e-06,
+      "loss": 0.9783,
+      "num_input_tokens_seen": 73779232,
+      "step": 147,
+      "train_runtime": 4144.4299,
+      "train_tokens_per_second": 17802.022
+    },
+    {
+      "epoch": 3.912751677852349,
+      "grad_norm": 1.041733980178833,
+      "learning_rate": 7.403709679352217e-06,
+      "loss": 0.3549,
+      "num_input_tokens_seen": 74283712,
+      "step": 148,
+      "train_runtime": 4170.0695,
+      "train_tokens_per_second": 17813.543
+    },
+    {
+      "epoch": 3.9395973154362416,
+      "grad_norm": 1.2669960260391235,
+      "learning_rate": 7.080437170788723e-06,
+      "loss": 0.303,
+      "num_input_tokens_seen": 74794496,
+      "step": 149,
+      "train_runtime": 4196.4068,
+      "train_tokens_per_second": 17823.462
+    },
+    {
+      "epoch": 3.966442953020134,
+      "grad_norm": 1.1353541612625122,
+      "learning_rate": 6.763212814534484e-06,
+      "loss": 0.3139,
+      "num_input_tokens_seen": 75296576,
+      "step": 150,
+      "train_runtime": 4221.9552,
+      "train_tokens_per_second": 17834.527
+    },
+    {
+      "epoch": 3.966442953020134,
+      "eval_accuracy": 0.8855249776173986,
+      "eval_loss": 0.42412105202674866,
+      "eval_runtime": 10.0868,
+      "eval_samples_per_second": 12.392,
+      "eval_steps_per_second": 0.793,
+      "num_input_tokens_seen": 75296576,
+      "step": 150
+    },
+    {
+      "epoch": 3.9932885906040267,
+      "grad_norm": 0.9825559258460999,
+      "learning_rate": 6.452143679117964e-06,
+      "loss": 1.9097,
+      "num_input_tokens_seen": 75810848,
+      "step": 151,
+      "train_runtime": 4318.6994,
+      "train_tokens_per_second": 17554.092
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.435434103012085,
+      "learning_rate": 6.147334755577596e-06,
+      "loss": 0.2221,
+      "num_input_tokens_seen": 75940992,
+      "step": 152,
+      "train_runtime": 4329.7093,
+      "train_tokens_per_second": 17539.513
+    },
+    {
+      "epoch": 4.026845637583893,
+      "grad_norm": 19.461166381835938,
+      "learning_rate": 5.848888922025553e-06,
+      "loss": 0.2591,
+      "num_input_tokens_seen": 76456096,
+      "step": 153,
+      "train_runtime": 4355.5057,
+      "train_tokens_per_second": 17553.896
+    },
+    {
+      "epoch": 4.053691275167785,
+      "grad_norm": 0.8980227708816528,
+      "learning_rate": 5.556906908924655e-06,
+      "loss": 0.2145,
+      "num_input_tokens_seen": 76975488,
+      "step": 154,
+      "train_runtime": 4381.4504,
+      "train_tokens_per_second": 17568.495
+    },
+    {
+      "epoch": 4.080536912751678,
+      "grad_norm": 0.8387576341629028,
+      "learning_rate": 5.271487265090163e-06,
+      "loss": 0.212,
+      "num_input_tokens_seen": 77487072,
+      "step": 155,
+      "train_runtime": 4407.6872,
+      "train_tokens_per_second": 17579.984
+    },
+    {
+      "epoch": 4.10738255033557,
+      "grad_norm": 0.9640564918518066,
+      "learning_rate": 4.992726324427901e-06,
+      "loss": 0.267,
+      "num_input_tokens_seen": 78007040,
+      "step": 156,
+      "train_runtime": 4433.5277,
+      "train_tokens_per_second": 17594.801
+    },
+    {
+      "epoch": 4.134228187919463,
+      "grad_norm": 0.8446484208106995,
+      "learning_rate": 4.720718173419947e-06,
+      "loss": 0.1877,
+      "num_input_tokens_seen": 78506528,
+      "step": 157,
+      "train_runtime": 4459.5141,
+      "train_tokens_per_second": 17604.279
+    },
+    {
+      "epoch": 4.1610738255033555,
+      "grad_norm": 41.02722930908203,
+      "learning_rate": 4.4555546193688735e-06,
+      "loss": 0.5,
+      "num_input_tokens_seen": 79019872,
+      "step": 158,
+      "train_runtime": 4485.9333,
+      "train_tokens_per_second": 17615.035
+    },
+    {
+      "epoch": 4.1879194630872485,
+      "grad_norm": 5.308436870574951,
+      "learning_rate": 4.19732515941125e-06,
+      "loss": 0.3085,
+      "num_input_tokens_seen": 79531008,
+      "step": 159,
+      "train_runtime": 4512.1558,
+      "train_tokens_per_second": 17625.944
+    },
+    {
+      "epoch": 4.214765100671141,
+      "grad_norm": 1.068027377128601,
+      "learning_rate": 3.94611695031086e-06,
+      "loss": 0.2403,
+      "num_input_tokens_seen": 80040288,
+      "step": 160,
+      "train_runtime": 4538.6362,
+      "train_tokens_per_second": 17635.317
+    },
+    {
+      "epoch": 4.241610738255034,
+      "grad_norm": 0.8074558973312378,
+      "learning_rate": 3.7020147790418263e-06,
+      "loss": 0.1888,
+      "num_input_tokens_seen": 80541216,
+      "step": 161,
+      "train_runtime": 4564.2981,
+      "train_tokens_per_second": 17645.915
+    },
+    {
+      "epoch": 4.268456375838926,
+      "grad_norm": 0.9346213936805725,
+      "learning_rate": 3.4651010341716028e-06,
+      "loss": 0.2003,
+      "num_input_tokens_seen": 81055936,
+      "step": 162,
+      "train_runtime": 4590.2865,
+      "train_tokens_per_second": 17658.143
+    },
+    {
+      "epoch": 4.295302013422819,
+      "grad_norm": 64.37608337402344,
+      "learning_rate": 3.2354556780534424e-06,
+      "loss": 1.3116,
+      "num_input_tokens_seen": 81567328,
+      "step": 163,
+      "train_runtime": 4616.1435,
+      "train_tokens_per_second": 17670.016
+    },
+    {
+      "epoch": 4.322147651006711,
+      "grad_norm": 1.8003010749816895,
+      "learning_rate": 3.013156219837776e-06,
+      "loss": 0.3799,
+      "num_input_tokens_seen": 82075904,
+      "step": 164,
+      "train_runtime": 4642.2255,
+      "train_tokens_per_second": 17680.292
+    },
+    {
+      "epoch": 4.348993288590604,
+      "grad_norm": 51.30678176879883,
+      "learning_rate": 2.7982776893115627e-06,
+      "loss": 0.8396,
+      "num_input_tokens_seen": 82579648,
+      "step": 165,
+      "train_runtime": 4667.8104,
+      "train_tokens_per_second": 17691.303
+    },
+    {
+      "epoch": 4.375838926174497,
+      "grad_norm": 0.9029057621955872,
+      "learning_rate": 2.5908926115744997e-06,
+      "loss": 0.2308,
+      "num_input_tokens_seen": 83089792,
+      "step": 166,
+      "train_runtime": 4693.866,
+      "train_tokens_per_second": 17701.782
+    },
+    {
+      "epoch": 4.402684563758389,
+      "grad_norm": 0.7935118675231934,
+      "learning_rate": 2.391070982560564e-06,
+      "loss": 0.2354,
+      "num_input_tokens_seen": 83575232,
+      "step": 167,
+      "train_runtime": 4719.8017,
+      "train_tokens_per_second": 17707.361
+    },
+    {
+      "epoch": 4.429530201342282,
+      "grad_norm": 0.7607171535491943,
+      "learning_rate": 2.1988802454132403e-06,
+      "loss": 0.1976,
+      "num_input_tokens_seen": 84087168,
+      "step": 168,
+      "train_runtime": 4746.1815,
+      "train_tokens_per_second": 17716.804
+    },
+    {
+      "epoch": 4.456375838926174,
+      "grad_norm": 0.8092939257621765,
+      "learning_rate": 2.0143852677223075e-06,
+      "loss": 0.1741,
+      "num_input_tokens_seen": 84598784,
+      "step": 169,
+      "train_runtime": 4772.2768,
+      "train_tokens_per_second": 17727.133
+    },
+    {
+      "epoch": 4.483221476510067,
+      "grad_norm": 0.8877170085906982,
+      "learning_rate": 1.837648319629956e-06,
+      "loss": 0.6518,
+      "num_input_tokens_seen": 85116288,
+      "step": 170,
+      "train_runtime": 4798.114,
+      "train_tokens_per_second": 17739.53
+    },
+    {
+      "epoch": 4.510067114093959,
+      "grad_norm": 0.7727687358856201,
+      "learning_rate": 1.6687290528135723e-06,
+      "loss": 0.2624,
+      "num_input_tokens_seen": 85632928,
+      "step": 171,
+      "train_runtime": 4824.2965,
+      "train_tokens_per_second": 17750.345
+    },
+    {
+      "epoch": 4.5369127516778525,
+      "grad_norm": 0.933302640914917,
+      "learning_rate": 1.5076844803522922e-06,
+      "loss": 0.27,
+      "num_input_tokens_seen": 86154432,
+      "step": 172,
+      "train_runtime": 4850.6581,
+      "train_tokens_per_second": 17761.39
+    },
+    {
+      "epoch": 4.563758389261745,
+      "grad_norm": 1.2586088180541992,
+      "learning_rate": 1.3545689574841342e-06,
+      "loss": 0.2335,
+      "num_input_tokens_seen": 86669984,
+      "step": 173,
+      "train_runtime": 4876.8647,
+      "train_tokens_per_second": 17771.66
+    },
+    {
+      "epoch": 4.590604026845638,
+      "grad_norm": 0.7908529043197632,
+      "learning_rate": 1.2094341632602064e-06,
+      "loss": 0.2193,
+      "num_input_tokens_seen": 87174880,
+      "step": 174,
+      "train_runtime": 4902.9706,
+      "train_tokens_per_second": 17780.013
+    },
+    {
+      "epoch": 4.617449664429531,
+      "grad_norm": 17.11946678161621,
+      "learning_rate": 1.0723290831021471e-06,
+      "loss": 0.6403,
+      "num_input_tokens_seen": 87699168,
+      "step": 175,
+      "train_runtime": 4929.4301,
+      "train_tokens_per_second": 17790.935
+    },
+    {
+      "epoch": 4.617449664429531,
+      "eval_accuracy": 0.8911537722889794,
+      "eval_loss": 0.4115161895751953,
+      "eval_runtime": 10.0488,
+      "eval_samples_per_second": 12.439,
+      "eval_steps_per_second": 0.796,
+      "num_input_tokens_seen": 87699168,
+      "step": 175
+    },
+    {
+      "epoch": 4.644295302013423,
+      "grad_norm": 2.2624847888946533,
+      "learning_rate": 9.432999922687396e-07,
+      "loss": 0.3758,
+      "num_input_tokens_seen": 88223360,
+      "step": 176,
+      "train_runtime": 5026.639,
+      "train_tokens_per_second": 17551.163
+    },
+    {
+      "epoch": 4.671140939597316,
+      "grad_norm": 0.6959147453308105,
+      "learning_rate": 8.223904402372334e-07,
+      "loss": 0.2342,
+      "num_input_tokens_seen": 88746912,
+      "step": 177,
+      "train_runtime": 5053.1227,
+      "train_tokens_per_second": 17562.786
+    },
+    {
+      "epoch": 4.697986577181208,
+      "grad_norm": 0.5389838814735413,
+      "learning_rate": 7.096412360046545e-07,
+      "loss": 0.2333,
+      "num_input_tokens_seen": 89264288,
+      "step": 178,
+      "train_runtime": 5079.4307,
+      "train_tokens_per_second": 17573.68
+    },
+    {
+      "epoch": 4.724832214765101,
+      "grad_norm": 5.69287109375,
+      "learning_rate": 6.050904343141095e-07,
+      "loss": 0.4271,
+      "num_input_tokens_seen": 89770272,
+      "step": 179,
+      "train_runtime": 5105.0048,
+      "train_tokens_per_second": 17584.758
+    },
+    {
+      "epoch": 4.751677852348993,
+      "grad_norm": 1.6489832401275635,
+      "learning_rate": 5.087733228106517e-07,
+      "loss": 0.3234,
+      "num_input_tokens_seen": 90293344,
+      "step": 180,
+      "train_runtime": 5131.3967,
+      "train_tokens_per_second": 17596.251
+    },
+    {
+      "epoch": 4.778523489932886,
+      "grad_norm": 6.22125768661499,
+      "learning_rate": 4.207224101311247e-07,
+      "loss": 0.2556,
+      "num_input_tokens_seen": 90813536,
+      "step": 181,
+      "train_runtime": 5157.5997,
+      "train_tokens_per_second": 17607.713
+    },
+    {
+      "epoch": 4.805369127516778,
+      "grad_norm": 0.8385538458824158,
+      "learning_rate": 3.4096741493194197e-07,
+      "loss": 0.1983,
+      "num_input_tokens_seen": 91300416,
+      "step": 182,
+      "train_runtime": 5183.0266,
+      "train_tokens_per_second": 17615.271
+    },
+    {
+      "epoch": 4.832214765100671,
+      "grad_norm": 1.0120928287506104,
+      "learning_rate": 2.6953525585855234e-07,
+      "loss": 0.2342,
+      "num_input_tokens_seen": 91817952,
+      "step": 183,
+      "train_runtime": 5208.8647,
+      "train_tokens_per_second": 17627.249
+    },
+    {
+      "epoch": 4.859060402684563,
+      "grad_norm": 1.149675965309143,
+      "learning_rate": 2.064500424599436e-07,
+      "loss": 0.2582,
+      "num_input_tokens_seen": 92328800,
+      "step": 184,
+      "train_runtime": 5234.9302,
+      "train_tokens_per_second": 17637.064
+    },
+    {
+      "epoch": 4.885906040268456,
+      "grad_norm": 1.0627394914627075,
+      "learning_rate": 1.517330670512629e-07,
+      "loss": 0.2364,
+      "num_input_tokens_seen": 92843328,
+      "step": 185,
+      "train_runtime": 5261.1199,
+      "train_tokens_per_second": 17647.066
+    },
+    {
+      "epoch": 4.912751677852349,
+      "grad_norm": 0.8081064820289612,
+      "learning_rate": 1.0540279752731253e-07,
+      "loss": 0.1979,
+      "num_input_tokens_seen": 93340480,
+      "step": 186,
+      "train_runtime": 5287.3612,
+      "train_tokens_per_second": 17653.509
+    },
+    {
+      "epoch": 4.939597315436242,
+      "grad_norm": 0.5786807537078857,
+      "learning_rate": 6.747487112931661e-08,
+      "loss": 0.6738,
+      "num_input_tokens_seen": 93856128,
+      "step": 187,
+      "train_runtime": 5314.1365,
+      "train_tokens_per_second": 17661.595
+    },
+    {
+      "epoch": 4.966442953020135,
+      "grad_norm": 0.6635648608207703,
+      "learning_rate": 3.796208916709565e-08,
+      "loss": 0.1781,
+      "num_input_tokens_seen": 94377760,
+      "step": 188,
+      "train_runtime": 5340.6451,
+      "train_tokens_per_second": 17671.603
+    },
+    {
+      "epoch": 4.993288590604027,
+      "grad_norm": 0.7493144273757935,
+      "learning_rate": 1.6874412698408836e-08,
+      "loss": 0.2207,
+      "num_input_tokens_seen": 94890496,
+      "step": 189,
+      "train_runtime": 5367.0696,
+      "train_tokens_per_second": 17680.131
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7493144273757935,
+      "learning_rate": 4.218959166932268e-09,
+      "loss": 0.3254,
+      "num_input_tokens_seen": 95021568,
+      "step": 190,
+      "train_runtime": 5372.17,
+      "train_tokens_per_second": 17687.744
+    },
+    {
+      "epoch": 5.0,
+      "num_input_tokens_seen": 95021568,
+      "step": 190,
+      "total_flos": 4.266242635254989e+18,
+      "train_loss": 0.7843870968410843,
+      "train_runtime": 5432.7802,
+      "train_samples_per_second": 2.186,
+      "train_steps_per_second": 0.035
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 190,
+  "num_input_tokens_seen": 95021568,
+  "num_train_epochs": 5,
+  "save_steps": 25,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.266242635254989e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}