初始化项目，由ModelHub XC社区提供模型

Model: Cisco1963/llmplasticity-zh_en_instant_0.25_1-seed42
Source: Original Platform

checkpoint-8/trainer_state.json

@@ -0,0 +1,98 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.0027934789725235156,
+  "eval_steps": 99999999,
+  "global_step": 8,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00034918487156543944,
+      "grad_norm": 232.1994171142578,
+      "learning_rate": 5e-05,
+      "loss": 175.5579,
+      "step": 1
+    },
+    {
+      "epoch": 0.00034918487156543944,
+      "eval_accuracy": 0.023954123581349192,
+      "eval_loss": 10.253963470458984,
+      "eval_runtime": 128.4941,
+      "eval_samples_per_second": 61.419,
+      "eval_steps_per_second": 7.681,
+      "step": 1
+    },
+    {
+      "epoch": 0.0006983697431308789,
+      "grad_norm": 114.17286682128906,
+      "learning_rate": 4.9982541899441346e-05,
+      "loss": 160.0191,
+      "step": 2
+    },
+    {
+      "epoch": 0.0006983697431308789,
+      "eval_accuracy": 0.016392362385855185,
+      "eval_loss": 9.9473237991333,
+      "eval_runtime": 128.9886,
+      "eval_samples_per_second": 61.184,
+      "eval_steps_per_second": 7.652,
+      "step": 2
+    },
+    {
+      "epoch": 0.0013967394862617578,
+      "grad_norm": 51.907196044921875,
+      "learning_rate": 4.9947625698324026e-05,
+      "loss": 152.5682,
+      "step": 4
+    },
+    {
+      "epoch": 0.0013967394862617578,
+      "eval_accuracy": 0.028750051402635483,
+      "eval_loss": 9.692176818847656,
+      "eval_runtime": 127.9133,
+      "eval_samples_per_second": 61.698,
+      "eval_steps_per_second": 7.716,
+      "step": 4
+    },
+    {
+      "epoch": 0.0027934789725235156,
+      "grad_norm": 79.11775970458984,
+      "learning_rate": 4.9877793296089386e-05,
+      "loss": 145.6449,
+      "step": 8
+    },
+    {
+      "epoch": 0.0027934789725235156,
+      "eval_accuracy": 0.1025968115998036,
+      "eval_loss": 10.163084030151367,
+      "eval_runtime": 156.5451,
+      "eval_samples_per_second": 50.414,
+      "eval_steps_per_second": 6.305,
+      "step": 8
+    }
+  ],
+  "logging_steps": 99999999,
+  "max_steps": 2864,
+  "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": 535126081536000.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}