Delete qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/README.md

@@ -1,61 +0,0 @@
----
-library_name: peft
-license: other
-base_model: /data/kcl/myt/Qwen2-7B-Instruct
-tags:
-- llama-factory
-- lora
-- generated_from_trainer
-model-index:
-- name: qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36
-  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. -->
-
-# qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36
-
-This model is a fine-tuned version of [/data/kcl/myt/Qwen2-7B-Instruct](https://huggingface.co//data/kcl/myt/Qwen2-7B-Instruct) on the SafeEdit_train dataset.
-
-## 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: 1e-05
-- train_batch_size: 2
-- eval_batch_size: 8
-- seed: 42
-- distributed_type: multi-GPU
-- num_devices: 4
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 64
-- total_eval_batch_size: 32
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- num_epochs: 3.0
-
-### Training results
-
-
-
-### Framework versions
-
-- PEFT 0.15.1
-- Transformers 4.45.0
-- Pytorch 2.6.0+cu124
-- Datasets 3.5.0
-- Tokenizers 0.20.3

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/adapter_config.json

@@ -1,39 +0,0 @@
-{
-  "alpha_pattern": {},
-  "auto_mapping": null,
-  "base_model_name_or_path": "/data/kcl/myt/Qwen2-7B-Instruct",
-  "bias": "none",
-  "corda_config": null,
-  "eva_config": null,
-  "exclude_modules": null,
-  "fan_in_fan_out": false,
-  "inference_mode": true,
-  "init_lora_weights": true,
-  "layer_replication": null,
-  "layers_pattern": null,
-  "layers_to_transform": null,
-  "loftq_config": {},
-  "lora_alpha": 16,
-  "lora_bias": false,
-  "lora_dropout": 0.0,
-  "megatron_config": null,
-  "megatron_core": "megatron.core",
-  "modules_to_save": null,
-  "peft_type": "LORA",
-  "r": 8,
-  "rank_pattern": {},
-  "revision": null,
-  "target_modules": [
-    "gate_proj",
-    "v_proj",
-    "q_proj",
-    "up_proj",
-    "o_proj",
-    "down_proj",
-    "k_proj"
-  ],
-  "task_type": "CAUSAL_LM",
-  "trainable_token_indices": null,
-  "use_dora": false,
-  "use_rslora": false
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/adapter_model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a831403938779bc5545eff2f88a521000c239dfde9a4906ddd92b844ce2579cf
-size 40422208

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/added_tokens.json

@@ -1,5 +0,0 @@
-{
-  "<|endoftext|>": 151643,
-  "<|im_end|>": 151645,
-  "<|im_start|>": 151644
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/all_results.json

@@ -1,9 +0,0 @@
-{
-    "epoch": 2.9822485207100593,
-    "num_input_tokens_seen": 6773184,
-    "total_flos": 316290046623744.0,
-    "train_loss": 1.2503274256590182,
-    "train_runtime": 2924.7101,
-    "train_samples_per_second": 4.154,
-    "train_steps_per_second": 0.065
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/special_tokens_map.json

@@ -1,20 +0,0 @@
-{
-  "additional_special_tokens": [
-    "<|im_start|>",
-    "<|im_end|>"
-  ],
-  "eos_token": {
-    "content": "<|im_end|>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
-  "pad_token": {
-    "content": "<|endoftext|>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  }
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/tokenizer.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:bcfe42da0a4497e8b2b172c1f9f4ec423a46dc12907f4349c55025f670422ba9
-size 11418266

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/tokenizer_config.json

@@ -1,44 +0,0 @@
-{
-  "add_prefix_space": false,
-  "added_tokens_decoder": {
-    "151643": {
-      "content": "<|endoftext|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "151644": {
-      "content": "<|im_start|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "151645": {
-      "content": "<|im_end|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    }
-  },
-  "additional_special_tokens": [
-    "<|im_start|>",
-    "<|im_end|>"
-  ],
-  "bos_token": null,
-  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
-  "clean_up_tokenization_spaces": false,
-  "eos_token": "<|im_end|>",
-  "errors": "replace",
-  "model_max_length": 131072,
-  "pad_token": "<|endoftext|>",
-  "padding_side": "right",
-  "split_special_tokens": false,
-  "tokenizer_class": "Qwen2Tokenizer",
-  "unk_token": null
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/train_results.json

@@ -1,9 +0,0 @@
-{
-    "epoch": 2.9822485207100593,
-    "num_input_tokens_seen": 6773184,
-    "total_flos": 316290046623744.0,
-    "train_loss": 1.2503274256590182,
-    "train_runtime": 2924.7101,
-    "train_samples_per_second": 4.154,
-    "train_steps_per_second": 0.065
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/trainer_log.jsonl

@@ -1,38 +0,0 @@
-{"current_steps": 5, "total_steps": 189, "loss": 1.6681, "lr": 9.982741376606077e-06, "epoch": 0.07889546351084813, "percentage": 2.65, "elapsed_time": "0:01:27", "remaining_time": "0:53:56", "throughput": 2031.39, "total_tokens": 178672}
-{"current_steps": 10, "total_steps": 189, "loss": 1.6603, "lr": 9.931084650456892e-06, "epoch": 0.15779092702169625, "percentage": 5.29, "elapsed_time": "0:02:43", "remaining_time": "0:48:40", "throughput": 2200.31, "total_tokens": 358944}
-{"current_steps": 15, "total_steps": 189, "loss": 1.6342, "lr": 9.84538643114539e-06, "epoch": 0.23668639053254437, "percentage": 7.94, "elapsed_time": "0:03:59", "remaining_time": "0:46:22", "throughput": 2245.31, "total_tokens": 538544}
-{"current_steps": 20, "total_steps": 189, "loss": 1.6051, "lr": 9.726238331988625e-06, "epoch": 0.3155818540433925, "percentage": 10.58, "elapsed_time": "0:05:14", "remaining_time": "0:44:20", "throughput": 2258.92, "total_tokens": 711280}
-{"current_steps": 25, "total_steps": 189, "loss": 1.5336, "lr": 9.574462885855173e-06, "epoch": 0.39447731755424065, "percentage": 13.23, "elapsed_time": "0:06:32", "remaining_time": "0:42:51", "throughput": 2271.47, "total_tokens": 890528}
-{"current_steps": 30, "total_steps": 189, "loss": 1.4808, "lr": 9.391107866851143e-06, "epoch": 0.47337278106508873, "percentage": 15.87, "elapsed_time": "0:07:46", "remaining_time": "0:41:13", "throughput": 2282.1, "total_tokens": 1064976}
-{"current_steps": 35, "total_steps": 189, "loss": 1.4176, "lr": 9.177439057064684e-06, "epoch": 0.5522682445759369, "percentage": 18.52, "elapsed_time": "0:09:03", "remaining_time": "0:39:49", "throughput": 2287.6, "total_tokens": 1242464}
-{"current_steps": 40, "total_steps": 189, "loss": 1.3607, "lr": 8.934931508303446e-06, "epoch": 0.631163708086785, "percentage": 21.16, "elapsed_time": "0:10:20", "remaining_time": "0:38:31", "throughput": 2296.65, "total_tokens": 1424864}
-{"current_steps": 45, "total_steps": 189, "loss": 1.3325, "lr": 8.665259359149132e-06, "epoch": 0.7100591715976331, "percentage": 23.81, "elapsed_time": "0:11:38", "remaining_time": "0:37:15", "throughput": 2297.52, "total_tokens": 1605216}
-{"current_steps": 50, "total_steps": 189, "loss": 1.3015, "lr": 8.370284277626576e-06, "epoch": 0.7889546351084813, "percentage": 26.46, "elapsed_time": "0:12:57", "remaining_time": "0:36:02", "throughput": 2298.03, "total_tokens": 1787600}
-{"current_steps": 55, "total_steps": 189, "loss": 1.2871, "lr": 8.052042609272817e-06, "epoch": 0.8678500986193294, "percentage": 29.1, "elapsed_time": "0:14:16", "remaining_time": "0:34:45", "throughput": 2301.16, "total_tokens": 1969856}
-{"current_steps": 60, "total_steps": 189, "loss": 1.2506, "lr": 7.712731319328798e-06, "epoch": 0.9467455621301775, "percentage": 31.75, "elapsed_time": "0:15:33", "remaining_time": "0:33:26", "throughput": 2304.85, "total_tokens": 2151056}
-{"current_steps": 65, "total_steps": 189, "loss": 1.2444, "lr": 7.354692826101102e-06, "epoch": 1.0256410256410255, "percentage": 34.39, "elapsed_time": "0:16:49", "remaining_time": "0:32:05", "throughput": 2311.48, "total_tokens": 2333280}
-{"current_steps": 70, "total_steps": 189, "loss": 1.2097, "lr": 6.980398830195785e-06, "epoch": 1.1045364891518739, "percentage": 37.04, "elapsed_time": "0:18:06", "remaining_time": "0:30:47", "throughput": 2313.91, "total_tokens": 2514720}
-{"current_steps": 75, "total_steps": 189, "loss": 1.183, "lr": 6.592433251258423e-06, "epoch": 1.183431952662722, "percentage": 39.68, "elapsed_time": "0:19:22", "remaining_time": "0:29:27", "throughput": 2314.84, "total_tokens": 2691392}
-{"current_steps": 80, "total_steps": 189, "loss": 1.2053, "lr": 6.19347439001569e-06, "epoch": 1.26232741617357, "percentage": 42.33, "elapsed_time": "0:20:40", "remaining_time": "0:28:10", "throughput": 2316.57, "total_tokens": 2874128}
-{"current_steps": 85, "total_steps": 189, "loss": 1.1725, "lr": 5.786276438761928e-06, "epoch": 1.3412228796844181, "percentage": 44.97, "elapsed_time": "0:21:57", "remaining_time": "0:26:51", "throughput": 2316.26, "total_tokens": 3051232}
-{"current_steps": 90, "total_steps": 189, "loss": 1.163, "lr": 5.373650467932122e-06, "epoch": 1.4201183431952662, "percentage": 47.62, "elapsed_time": "0:23:17", "remaining_time": "0:25:36", "throughput": 2315.9, "total_tokens": 3235360}
-{"current_steps": 95, "total_steps": 189, "loss": 1.1811, "lr": 4.958445020019516e-06, "epoch": 1.4990138067061145, "percentage": 50.26, "elapsed_time": "0:24:34", "remaining_time": "0:24:18", "throughput": 2317.33, "total_tokens": 3416864}
-{"current_steps": 100, "total_steps": 189, "loss": 1.1796, "lr": 4.5435264448067595e-06, "epoch": 1.5779092702169626, "percentage": 52.91, "elapsed_time": "0:25:50", "remaining_time": "0:22:59", "throughput": 2315.15, "total_tokens": 3588656}
-{"current_steps": 105, "total_steps": 189, "loss": 1.1236, "lr": 4.131759111665349e-06, "epoch": 1.6568047337278107, "percentage": 55.56, "elapsed_time": "0:27:18", "remaining_time": "0:21:51", "throughput": 2300.51, "total_tokens": 3770480}
-{"current_steps": 110, "total_steps": 189, "loss": 1.14, "lr": 3.7259856355267275e-06, "epoch": 1.7357001972386588, "percentage": 58.2, "elapsed_time": "0:28:33", "remaining_time": "0:20:30", "throughput": 2305.43, "total_tokens": 3949808}
-{"current_steps": 115, "total_steps": 189, "loss": 1.1436, "lr": 3.3290072530340628e-06, "epoch": 1.8145956607495068, "percentage": 60.85, "elapsed_time": "0:29:49", "remaining_time": "0:19:11", "throughput": 2308.29, "total_tokens": 4129552}
-{"current_steps": 120, "total_steps": 189, "loss": 1.1168, "lr": 2.9435644843469434e-06, "epoch": 1.893491124260355, "percentage": 63.49, "elapsed_time": "0:31:05", "remaining_time": "0:17:52", "throughput": 2309.33, "total_tokens": 4307456}
-{"current_steps": 125, "total_steps": 189, "loss": 1.183, "lr": 2.5723182140992385e-06, "epoch": 1.972386587771203, "percentage": 66.14, "elapsed_time": "0:32:20", "remaining_time": "0:16:33", "throughput": 2310.49, "total_tokens": 4482576}
-{"current_steps": 130, "total_steps": 189, "loss": 1.1061, "lr": 2.217831322116797e-06, "epoch": 2.051282051282051, "percentage": 68.78, "elapsed_time": "0:33:36", "remaining_time": "0:15:15", "throughput": 2313.13, "total_tokens": 4664000}
-{"current_steps": 135, "total_steps": 189, "loss": 1.1299, "lr": 1.8825509907063328e-06, "epoch": 2.1301775147928996, "percentage": 71.43, "elapsed_time": "0:34:50", "remaining_time": "0:13:56", "throughput": 2315.18, "total_tokens": 4840624}
-{"current_steps": 140, "total_steps": 189, "loss": 1.1334, "lr": 1.5687918106563326e-06, "epoch": 2.2090729783037477, "percentage": 74.07, "elapsed_time": "0:36:07", "remaining_time": "0:12:38", "throughput": 2315.32, "total_tokens": 5019584}
-{"current_steps": 145, "total_steps": 189, "loss": 1.1393, "lr": 1.2787198025767417e-06, "epoch": 2.287968441814596, "percentage": 76.72, "elapsed_time": "0:37:25", "remaining_time": "0:11:21", "throughput": 2316.11, "total_tokens": 5199712}
-{"current_steps": 150, "total_steps": 189, "loss": 1.1358, "lr": 1.0143374638853892e-06, "epoch": 2.366863905325444, "percentage": 79.37, "elapsed_time": "0:38:41", "remaining_time": "0:10:03", "throughput": 2318.52, "total_tokens": 5381712}
-{"current_steps": 155, "total_steps": 189, "loss": 1.1423, "lr": 7.774699446684608e-07, "epoch": 2.445759368836292, "percentage": 82.01, "elapsed_time": "0:39:58", "remaining_time": "0:08:46", "throughput": 2320.76, "total_tokens": 5565648}
-{"current_steps": 160, "total_steps": 189, "loss": 1.1349, "lr": 5.697524478493288e-07, "epoch": 2.52465483234714, "percentage": 84.66, "elapsed_time": "0:41:13", "remaining_time": "0:07:28", "throughput": 2323.18, "total_tokens": 5745664}
-{"current_steps": 165, "total_steps": 189, "loss": 1.1684, "lr": 3.9261894064796136e-07, "epoch": 2.603550295857988, "percentage": 87.3, "elapsed_time": "0:42:32", "remaining_time": "0:06:11", "throughput": 2321.49, "total_tokens": 5924448}
-{"current_steps": 170, "total_steps": 189, "loss": 1.1034, "lr": 2.472922552608559e-07, "epoch": 2.6824457593688362, "percentage": 89.95, "elapsed_time": "0:43:46", "remaining_time": "0:04:53", "throughput": 2323.57, "total_tokens": 6102352}
-{"current_steps": 175, "total_steps": 189, "loss": 1.1398, "lr": 1.3477564710088097e-07, "epoch": 2.7613412228796843, "percentage": 92.59, "elapsed_time": "0:45:01", "remaining_time": "0:03:36", "throughput": 2323.27, "total_tokens": 6276448}
-{"current_steps": 180, "total_steps": 189, "loss": 1.1105, "lr": 5.584586887435739e-08, "epoch": 2.8402366863905324, "percentage": 95.24, "elapsed_time": "0:46:17", "remaining_time": "0:02:18", "throughput": 2323.62, "total_tokens": 6454720}
-{"current_steps": 185, "total_steps": 189, "loss": 1.1305, "lr": 1.1047808308075059e-08, "epoch": 2.9191321499013805, "percentage": 97.88, "elapsed_time": "0:47:34", "remaining_time": "0:01:01", "throughput": 2324.4, "total_tokens": 6634464}
-{"current_steps": 189, "total_steps": 189, "epoch": 2.9822485207100593, "percentage": 100.0, "elapsed_time": "0:48:44", "remaining_time": "0:00:00", "throughput": 2315.84, "total_tokens": 6773184}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/trainer_state.json

@@ -1,339 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 2.9822485207100593,
-  "eval_steps": 500,
-  "global_step": 189,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.07889546351084813,
-      "grad_norm": 0.7517743802879164,
-      "learning_rate": 9.982741376606077e-06,
-      "loss": 1.6681,
-      "num_input_tokens_seen": 178672,
-      "step": 5
-    },
-    {
-      "epoch": 0.15779092702169625,
-      "grad_norm": 0.844390565883667,
-      "learning_rate": 9.931084650456892e-06,
-      "loss": 1.6603,
-      "num_input_tokens_seen": 358944,
-      "step": 10
-    },
-    {
-      "epoch": 0.23668639053254437,
-      "grad_norm": 0.9001854481599434,
-      "learning_rate": 9.84538643114539e-06,
-      "loss": 1.6342,
-      "num_input_tokens_seen": 538544,
-      "step": 15
-    },
-    {
-      "epoch": 0.3155818540433925,
-      "grad_norm": 0.9052489549925832,
-      "learning_rate": 9.726238331988625e-06,
-      "loss": 1.6051,
-      "num_input_tokens_seen": 711280,
-      "step": 20
-    },
-    {
-      "epoch": 0.39447731755424065,
-      "grad_norm": 0.8027376292951401,
-      "learning_rate": 9.574462885855173e-06,
-      "loss": 1.5336,
-      "num_input_tokens_seen": 890528,
-      "step": 25
-    },
-    {
-      "epoch": 0.47337278106508873,
-      "grad_norm": 0.7870518685476616,
-      "learning_rate": 9.391107866851143e-06,
-      "loss": 1.4808,
-      "num_input_tokens_seen": 1064976,
-      "step": 30
-    },
-    {
-      "epoch": 0.5522682445759369,
-      "grad_norm": 0.6854923958247626,
-      "learning_rate": 9.177439057064684e-06,
-      "loss": 1.4176,
-      "num_input_tokens_seen": 1242464,
-      "step": 35
-    },
-    {
-      "epoch": 0.631163708086785,
-      "grad_norm": 0.5750854251722598,
-      "learning_rate": 8.934931508303446e-06,
-      "loss": 1.3607,
-      "num_input_tokens_seen": 1424864,
-      "step": 40
-    },
-    {
-      "epoch": 0.7100591715976331,
-      "grad_norm": 0.5472056176391153,
-      "learning_rate": 8.665259359149132e-06,
-      "loss": 1.3325,
-      "num_input_tokens_seen": 1605216,
-      "step": 45
-    },
-    {
-      "epoch": 0.7889546351084813,
-      "grad_norm": 0.48469354587632074,
-      "learning_rate": 8.370284277626576e-06,
-      "loss": 1.3015,
-      "num_input_tokens_seen": 1787600,
-      "step": 50
-    },
-    {
-      "epoch": 0.8678500986193294,
-      "grad_norm": 0.5100775865785815,
-      "learning_rate": 8.052042609272817e-06,
-      "loss": 1.2871,
-      "num_input_tokens_seen": 1969856,
-      "step": 55
-    },
-    {
-      "epoch": 0.9467455621301775,
-      "grad_norm": 0.4278394592846904,
-      "learning_rate": 7.712731319328798e-06,
-      "loss": 1.2506,
-      "num_input_tokens_seen": 2151056,
-      "step": 60
-    },
-    {
-      "epoch": 1.0256410256410255,
-      "grad_norm": 0.4445611249536644,
-      "learning_rate": 7.354692826101102e-06,
-      "loss": 1.2444,
-      "num_input_tokens_seen": 2333280,
-      "step": 65
-    },
-    {
-      "epoch": 1.1045364891518739,
-      "grad_norm": 0.4249772475239759,
-      "learning_rate": 6.980398830195785e-06,
-      "loss": 1.2097,
-      "num_input_tokens_seen": 2514720,
-      "step": 70
-    },
-    {
-      "epoch": 1.183431952662722,
-      "grad_norm": 0.3890283360722578,
-      "learning_rate": 6.592433251258423e-06,
-      "loss": 1.183,
-      "num_input_tokens_seen": 2691392,
-      "step": 75
-    },
-    {
-      "epoch": 1.26232741617357,
-      "grad_norm": 0.38495737090738175,
-      "learning_rate": 6.19347439001569e-06,
-      "loss": 1.2053,
-      "num_input_tokens_seen": 2874128,
-      "step": 80
-    },
-    {
-      "epoch": 1.3412228796844181,
-      "grad_norm": 0.3810624778090492,
-      "learning_rate": 5.786276438761928e-06,
-      "loss": 1.1725,
-      "num_input_tokens_seen": 3051232,
-      "step": 85
-    },
-    {
-      "epoch": 1.4201183431952662,
-      "grad_norm": 0.38362267333836025,
-      "learning_rate": 5.373650467932122e-06,
-      "loss": 1.163,
-      "num_input_tokens_seen": 3235360,
-      "step": 90
-    },
-    {
-      "epoch": 1.4990138067061145,
-      "grad_norm": 0.368418324987837,
-      "learning_rate": 4.958445020019516e-06,
-      "loss": 1.1811,
-      "num_input_tokens_seen": 3416864,
-      "step": 95
-    },
-    {
-      "epoch": 1.5779092702169626,
-      "grad_norm": 0.383407588436673,
-      "learning_rate": 4.5435264448067595e-06,
-      "loss": 1.1796,
-      "num_input_tokens_seen": 3588656,
-      "step": 100
-    },
-    {
-      "epoch": 1.6568047337278107,
-      "grad_norm": 0.35574670619233356,
-      "learning_rate": 4.131759111665349e-06,
-      "loss": 1.1236,
-      "num_input_tokens_seen": 3770480,
-      "step": 105
-    },
-    {
-      "epoch": 1.7357001972386588,
-      "grad_norm": 0.37423484457216416,
-      "learning_rate": 3.7259856355267275e-06,
-      "loss": 1.14,
-      "num_input_tokens_seen": 3949808,
-      "step": 110
-    },
-    {
-      "epoch": 1.8145956607495068,
-      "grad_norm": 0.36784958426111003,
-      "learning_rate": 3.3290072530340628e-06,
-      "loss": 1.1436,
-      "num_input_tokens_seen": 4129552,
-      "step": 115
-    },
-    {
-      "epoch": 1.893491124260355,
-      "grad_norm": 0.34472939740012404,
-      "learning_rate": 2.9435644843469434e-06,
-      "loss": 1.1168,
-      "num_input_tokens_seen": 4307456,
-      "step": 120
-    },
-    {
-      "epoch": 1.972386587771203,
-      "grad_norm": 0.3888969617469796,
-      "learning_rate": 2.5723182140992385e-06,
-      "loss": 1.183,
-      "num_input_tokens_seen": 4482576,
-      "step": 125
-    },
-    {
-      "epoch": 2.051282051282051,
-      "grad_norm": 0.350147395425708,
-      "learning_rate": 2.217831322116797e-06,
-      "loss": 1.1061,
-      "num_input_tokens_seen": 4664000,
-      "step": 130
-    },
-    {
-      "epoch": 2.1301775147928996,
-      "grad_norm": 0.3861518755075231,
-      "learning_rate": 1.8825509907063328e-06,
-      "loss": 1.1299,
-      "num_input_tokens_seen": 4840624,
-      "step": 135
-    },
-    {
-      "epoch": 2.2090729783037477,
-      "grad_norm": 0.3546443595857142,
-      "learning_rate": 1.5687918106563326e-06,
-      "loss": 1.1334,
-      "num_input_tokens_seen": 5019584,
-      "step": 140
-    },
-    {
-      "epoch": 2.287968441814596,
-      "grad_norm": 0.34227747301098843,
-      "learning_rate": 1.2787198025767417e-06,
-      "loss": 1.1393,
-      "num_input_tokens_seen": 5199712,
-      "step": 145
-    },
-    {
-      "epoch": 2.366863905325444,
-      "grad_norm": 0.3394440679416029,
-      "learning_rate": 1.0143374638853892e-06,
-      "loss": 1.1358,
-      "num_input_tokens_seen": 5381712,
-      "step": 150
-    },
-    {
-      "epoch": 2.445759368836292,
-      "grad_norm": 0.3437394931824312,
-      "learning_rate": 7.774699446684608e-07,
-      "loss": 1.1423,
-      "num_input_tokens_seen": 5565648,
-      "step": 155
-    },
-    {
-      "epoch": 2.52465483234714,
-      "grad_norm": 0.34664670822827176,
-      "learning_rate": 5.697524478493288e-07,
-      "loss": 1.1349,
-      "num_input_tokens_seen": 5745664,
-      "step": 160
-    },
-    {
-      "epoch": 2.603550295857988,
-      "grad_norm": 0.33354356146833053,
-      "learning_rate": 3.9261894064796136e-07,
-      "loss": 1.1684,
-      "num_input_tokens_seen": 5924448,
-      "step": 165
-    },
-    {
-      "epoch": 2.6824457593688362,
-      "grad_norm": 0.37735672802246767,
-      "learning_rate": 2.472922552608559e-07,
-      "loss": 1.1034,
-      "num_input_tokens_seen": 6102352,
-      "step": 170
-    },
-    {
-      "epoch": 2.7613412228796843,
-      "grad_norm": 0.36263778412439085,
-      "learning_rate": 1.3477564710088097e-07,
-      "loss": 1.1398,
-      "num_input_tokens_seen": 6276448,
-      "step": 175
-    },
-    {
-      "epoch": 2.8402366863905324,
-      "grad_norm": 0.3793915300588604,
-      "learning_rate": 5.584586887435739e-08,
-      "loss": 1.1105,
-      "num_input_tokens_seen": 6454720,
-      "step": 180
-    },
-    {
-      "epoch": 2.9191321499013805,
-      "grad_norm": 0.34271636031236846,
-      "learning_rate": 1.1047808308075059e-08,
-      "loss": 1.1305,
-      "num_input_tokens_seen": 6634464,
-      "step": 185
-    },
-    {
-      "epoch": 2.9822485207100593,
-      "num_input_tokens_seen": 6773184,
-      "step": 189,
-      "total_flos": 316290046623744.0,
-      "train_loss": 1.2503274256590182,
-      "train_runtime": 2924.7101,
-      "train_samples_per_second": 4.154,
-      "train_steps_per_second": 0.065
-    }
-  ],
-  "logging_steps": 5,
-  "max_steps": 189,
-  "num_input_tokens_seen": 6773184,
-  "num_train_epochs": 3,
-  "save_steps": 100,
-  "stateful_callbacks": {
-    "TrainerControl": {
-      "args": {
-        "should_epoch_stop": false,
-        "should_evaluate": false,
-        "should_log": false,
-        "should_save": true,
-        "should_training_stop": true
-      },
-      "attributes": {}
-    }
-  },
-  "total_flos": 316290046623744.0,
-  "train_batch_size": 2,
-  "trial_name": null,
-  "trial_params": null
-}

qwen2-7b-instruct_refusal-sft-lora_v1_2025-09-26-03-11-36/training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:578d7264005d7bb2415375feb4d2afc32a3ca447e14ffc34c1b7fef987277d58
-size 7480