Training in progress, step 9920, checkpoint

checkpoint-9920/config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "pad_token_id": 2,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 30000
+}

checkpoint-9920/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "pad_token_id": 2,
+  "transformers_version": "4.56.1"
+}

checkpoint-9920/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5323a059e7b5240aecb6568d18d9f54419d9823c35f3aa5d027b7e2bac4b740c
+size 435544704

checkpoint-9920/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a46f033f378f1c7c52e43e749fc93b365cdc761453b6e319adaf7b8108af3979
+size 871183627

checkpoint-9920/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1805f983f2f2a36dd83544a132bc9cd46274278f4cc440b543543e1e104d7eea
+size 14709

checkpoint-9920/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:297e84f888348f187e5937b586f8013b1b07998e4f34900f2d0e28e31e1491ad
+size 1383

checkpoint-9920/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:541d5659420db828b0d58735fe059c21ed958108511a66f46c178b5afe1cce0e
+size 1465

checkpoint-9920/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

checkpoint-9920/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-9920/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 4960,
+  "best_metric": 3.0957539081573486,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 9920,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0020161290322580645,
+      "grad_norm": 27.16815185546875,
+      "learning_rate": 0.0,
+      "loss": 10.3826,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.582794189453125,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 4.588,
+      "step": 496
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.528367042541504,
+      "eval_runtime": 19.7121,
+      "eval_samples_per_second": 1615.103,
+      "eval_steps_per_second": 6.341,
+      "step": 496
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.8563077449798584,
+      "learning_rate": 9.477659574468086e-05,
+      "loss": 3.3226,
+      "step": 992
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.304581642150879,
+      "eval_runtime": 19.6517,
+      "eval_samples_per_second": 1620.064,
+      "eval_steps_per_second": 6.361,
+      "step": 992
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.8085999488830566,
+      "learning_rate": 8.950000000000001e-05,
+      "loss": 3.1682,
+      "step": 1488
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.2201364040374756,
+      "eval_runtime": 19.7315,
+      "eval_samples_per_second": 1613.513,
+      "eval_steps_per_second": 6.335,
+      "step": 1488
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.040928840637207,
+      "learning_rate": 8.422340425531914e-05,
+      "loss": 3.0809,
+      "step": 1984
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.1738715171813965,
+      "eval_runtime": 19.9591,
+      "eval_samples_per_second": 1595.111,
+      "eval_steps_per_second": 6.263,
+      "step": 1984
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 4.117753028869629,
+      "learning_rate": 7.894680851063831e-05,
+      "loss": 3.0157,
+      "step": 2480
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.1431686878204346,
+      "eval_runtime": 19.7234,
+      "eval_samples_per_second": 1614.172,
+      "eval_steps_per_second": 6.338,
+      "step": 2480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.192259788513184,
+      "learning_rate": 7.367021276595744e-05,
+      "loss": 2.9617,
+      "step": 2976
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.123141050338745,
+      "eval_runtime": 19.7316,
+      "eval_samples_per_second": 1613.507,
+      "eval_steps_per_second": 6.335,
+      "step": 2976
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.601728916168213,
+      "learning_rate": 6.83936170212766e-05,
+      "loss": 2.9134,
+      "step": 3472
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.110758066177368,
+      "eval_runtime": 19.7226,
+      "eval_samples_per_second": 1614.243,
+      "eval_steps_per_second": 6.338,
+      "step": 3472
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.684045314788818,
+      "learning_rate": 6.311702127659575e-05,
+      "loss": 2.8706,
+      "step": 3968
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.1021018028259277,
+      "eval_runtime": 19.9219,
+      "eval_samples_per_second": 1598.088,
+      "eval_steps_per_second": 6.274,
+      "step": 3968
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.68789005279541,
+      "learning_rate": 5.7840425531914896e-05,
+      "loss": 2.828,
+      "step": 4464
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.0959761142730713,
+      "eval_runtime": 19.7592,
+      "eval_samples_per_second": 1611.251,
+      "eval_steps_per_second": 6.326,
+      "step": 4464
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 5.364582538604736,
+      "learning_rate": 5.256382978723404e-05,
+      "loss": 2.7883,
+      "step": 4960
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.0957539081573486,
+      "eval_runtime": 19.8171,
+      "eval_samples_per_second": 1606.538,
+      "eval_steps_per_second": 6.308,
+      "step": 4960
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 5.790548324584961,
+      "learning_rate": 4.728723404255319e-05,
+      "loss": 2.7484,
+      "step": 5456
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.0989372730255127,
+      "eval_runtime": 19.7155,
+      "eval_samples_per_second": 1614.822,
+      "eval_steps_per_second": 6.34,
+      "step": 5456
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 5.615660190582275,
+      "learning_rate": 4.201063829787234e-05,
+      "loss": 2.7102,
+      "step": 5952
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.1046364307403564,
+      "eval_runtime": 19.7412,
+      "eval_samples_per_second": 1612.722,
+      "eval_steps_per_second": 6.332,
+      "step": 5952
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 7.992080211639404,
+      "learning_rate": 3.673404255319149e-05,
+      "loss": 2.6743,
+      "step": 6448
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.1134583950042725,
+      "eval_runtime": 19.718,
+      "eval_samples_per_second": 1614.612,
+      "eval_steps_per_second": 6.339,
+      "step": 6448
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 6.8963847160339355,
+      "learning_rate": 3.1457446808510635e-05,
+      "loss": 2.6381,
+      "step": 6944
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.1234219074249268,
+      "eval_runtime": 19.7214,
+      "eval_samples_per_second": 1614.34,
+      "eval_steps_per_second": 6.338,
+      "step": 6944
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 8.328634262084961,
+      "learning_rate": 2.6180851063829785e-05,
+      "loss": 2.6059,
+      "step": 7440
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.1336703300476074,
+      "eval_runtime": 19.723,
+      "eval_samples_per_second": 1614.207,
+      "eval_steps_per_second": 6.338,
+      "step": 7440
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 6.984272003173828,
+      "learning_rate": 2.090425531914894e-05,
+      "loss": 2.5728,
+      "step": 7936
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.1444923877716064,
+      "eval_runtime": 19.7056,
+      "eval_samples_per_second": 1615.635,
+      "eval_steps_per_second": 6.343,
+      "step": 7936
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": Infinity,
+      "learning_rate": 1.5627659574468086e-05,
+      "loss": 2.5441,
+      "step": 8432
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.156653881072998,
+      "eval_runtime": 19.7474,
+      "eval_samples_per_second": 1612.213,
+      "eval_steps_per_second": 6.33,
+      "step": 8432
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": Infinity,
+      "learning_rate": 1.0361702127659576e-05,
+      "loss": 2.5208,
+      "step": 8928
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 3.164964437484741,
+      "eval_runtime": 19.7475,
+      "eval_samples_per_second": 1612.204,
+      "eval_steps_per_second": 6.33,
+      "step": 8928
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 9.182657241821289,
+      "learning_rate": 5.095744680851064e-06,
+      "loss": 2.5005,
+      "step": 9424
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 3.169865369796753,
+      "eval_runtime": 19.7663,
+      "eval_samples_per_second": 1610.671,
+      "eval_steps_per_second": 6.324,
+      "step": 9424
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9920,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.6555985731584e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-9920/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5181755f3137a55d46907e62f212bba95eb37f3acebf805ca09cb911d4a30e8
+size 5905