初始化项目，由ModelHub XC社区提供模型

Model: Cisco1963/llmplasticity-de_en_linear_0.5_1-seed42
Source: Original Platform

checkpoint-256/config.json

@@ -0,0 +1,39 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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": 50256,
+  "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,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "transformers_version": "4.57.1",
+  "use_cache": true,
+  "vocab_size": 50257
+}

checkpoint-256/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.57.1"
+}

checkpoint-256/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72a506053b94fc292eeaadd9fca9f62d3f3716533a655385cbec4c0ff6d6f458
+size 497774208

checkpoint-256/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

checkpoint-256/tokenizer_config.json

@@ -0,0 +1,21 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

checkpoint-256/trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.08029798078808077,
+  "eval_steps": 99999999,
+  "global_step": 256,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0003136639874534405,
+      "grad_norm": 174.2938690185547,
+      "learning_rate": 5e-05,
+      "loss": 175.3938,
+      "step": 1
+    },
+    {
+      "epoch": 0.0003136639874534405,
+      "eval_accuracy": 0.038034240719819076,
+      "eval_loss": 10.384196281433105,
+      "eval_runtime": 194.5696,
+      "eval_samples_per_second": 29.83,
+      "eval_steps_per_second": 3.731,
+      "step": 1
+    },
+    {
+      "epoch": 0.000627327974906881,
+      "grad_norm": 83.63700103759766,
+      "learning_rate": 4.99843211037943e-05,
+      "loss": 164.5011,
+      "step": 2
+    },
+    {
+      "epoch": 0.000627327974906881,
+      "eval_accuracy": 0.038057314435118395,
+      "eval_loss": 10.165145874023438,
+      "eval_runtime": 259.579,
+      "eval_samples_per_second": 22.359,
+      "eval_steps_per_second": 2.797,
+      "step": 2
+    },
+    {
+      "epoch": 0.001254655949813762,
+      "grad_norm": 43.825782775878906,
+      "learning_rate": 4.9952963311382884e-05,
+      "loss": 158.7612,
+      "step": 4
+    },
+    {
+      "epoch": 0.001254655949813762,
+      "eval_accuracy": 0.038057314435118395,
+      "eval_loss": 10.000017166137695,
+      "eval_runtime": 258.7681,
+      "eval_samples_per_second": 22.429,
+      "eval_steps_per_second": 2.806,
+      "step": 4
+    },
+    {
+      "epoch": 0.002509311899627524,
+      "grad_norm": 35.26978302001953,
+      "learning_rate": 4.9890247726560055e-05,
+      "loss": 155.0138,
+      "step": 8
+    },
+    {
+      "epoch": 0.002509311899627524,
+      "eval_accuracy": 0.0483111387771133,
+      "eval_loss": 9.72864055633545,
+      "eval_runtime": 309.1294,
+      "eval_samples_per_second": 18.775,
+      "eval_steps_per_second": 2.349,
+      "step": 8
+    },
+    {
+      "epoch": 0.005018623799255048,
+      "grad_norm": 40.16830825805664,
+      "learning_rate": 4.97648165569144e-05,
+      "loss": 149.2228,
+      "step": 16
+    },
+    {
+      "epoch": 0.005018623799255048,
+      "eval_accuracy": 0.033622108459261925,
+      "eval_loss": 11.213191986083984,
+      "eval_runtime": 112.2816,
+      "eval_samples_per_second": 51.691,
+      "eval_steps_per_second": 6.466,
+      "step": 16
+    },
+    {
+      "epoch": 0.010037247598510097,
+      "grad_norm": 37.20323944091797,
+      "learning_rate": 4.951395421762308e-05,
+      "loss": 145.2656,
+      "step": 32
+    },
+    {
+      "epoch": 0.010037247598510097,
+      "eval_accuracy": 0.06755377523034978,
+      "eval_loss": 8.518936157226562,
+      "eval_runtime": 113.0533,
+      "eval_samples_per_second": 51.339,
+      "eval_steps_per_second": 6.422,
+      "step": 32
+    },
+    {
+      "epoch": 0.020074495197020193,
+      "grad_norm": 24.089946746826172,
+      "learning_rate": 4.9012229539040455e-05,
+      "loss": 115.6545,
+      "step": 64
+    },
+    {
+      "epoch": 0.020074495197020193,
+      "eval_accuracy": 0.0989353754076637,
+      "eval_loss": 7.334339618682861,
+      "eval_runtime": 118.0378,
+      "eval_samples_per_second": 49.171,
+      "eval_steps_per_second": 6.151,
+      "step": 64
+    },
+    {
+      "epoch": 0.040148990394040386,
+      "grad_norm": 13.746461868286133,
+      "learning_rate": 4.80087801818752e-05,
+      "loss": 93.3162,
+      "step": 128
+    },
+    {
+      "epoch": 0.040148990394040386,
+      "eval_accuracy": 0.12674374971789434,
+      "eval_loss": 6.555325508117676,
+      "eval_runtime": 111.1576,
+      "eval_samples_per_second": 52.214,
+      "eval_steps_per_second": 6.531,
+      "step": 128
+    },
+    {
+      "epoch": 0.08029798078808077,
+      "grad_norm": 13.765669822692871,
+      "learning_rate": 4.600188146754469e-05,
+      "loss": 77.7084,
+      "step": 256
+    },
+    {
+      "epoch": 0.08029798078808077,
+      "eval_accuracy": 0.16966187070231,
+      "eval_loss": 5.74245548248291,
+      "eval_runtime": 112.901,
+      "eval_samples_per_second": 51.408,
+      "eval_steps_per_second": 6.43,
+      "step": 256
+    }
+  ],
+  "logging_steps": 99999999,
+  "max_steps": 3189,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 99999999,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.7124034609152e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-256/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6e1105b0d5c6e05867600af834e2f856bce3d05846453054669896adc127ca6
+size 5969