初始化项目，由ModelHub XC社区提供模型

Model: voidful/Qwen3-0.6B-SFT-Tulu3-FD
Source: Original Platform

checkpoint-15/trainer_state.json

@@ -0,0 +1,44 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 15,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 0.9494459543377161,
+      "epoch": 0.6666666666666666,
+      "grad_norm": 1.175722360610962,
+      "learning_rate": 3.887395330218429e-06,
+      "loss": 0.976,
+      "mean_token_accuracy": 0.764532539434731,
+      "num_tokens": 20489220.0,
+      "step": 10
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 15,
+  "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": 8.160891840443187e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}