初始化项目，由ModelHub XC社区提供模型

Model: FlyPig23/Llama3.2-3B_Paper_Impact_dataset_SFT_1ep
Source: Original Platform

trainer_state.json

@@ -0,0 +1,85 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 34,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.14925373134328357,
+      "grad_norm": 1.9213414192199707,
+      "learning_rate": 2e-05,
+      "loss": 0.1823,
+      "step": 5
+    },
+    {
+      "epoch": 0.29850746268656714,
+      "grad_norm": 0.25910183787345886,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.1069,
+      "step": 10
+    },
+    {
+      "epoch": 0.44776119402985076,
+      "grad_norm": 0.047786250710487366,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.1004,
+      "step": 15
+    },
+    {
+      "epoch": 0.5970149253731343,
+      "grad_norm": 0.009523790329694748,
+      "learning_rate": 1e-05,
+      "loss": 0.1036,
+      "step": 20
+    },
+    {
+      "epoch": 0.746268656716418,
+      "grad_norm": 0.021309392526745796,
+      "learning_rate": 5.000000000000003e-06,
+      "loss": 0.1,
+      "step": 25
+    },
+    {
+      "epoch": 0.8955223880597015,
+      "grad_norm": 0.020632434636354446,
+      "learning_rate": 1.339745962155613e-06,
+      "loss": 0.0983,
+      "step": 30
+    },
+    {
+      "epoch": 1.0,
+      "step": 34,
+      "total_flos": 1.8521361846213018e+17,
+      "train_loss": 0.11334303547354306,
+      "train_runtime": 974.5943,
+      "train_samples_per_second": 4.38,
+      "train_steps_per_second": 0.035
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 34,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "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": 1.8521361846213018e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}