初始化项目，由ModelHub XC社区提供模型

Model: laion/sera-316-opt1k__Qwen3-8B
Source: Original Platform

trainer_state.json

@@ -0,0 +1,98 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 28,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.3,
+      "grad_norm": 0.770761240772315,
+      "learning_rate": 3.9842294026289565e-05,
+      "loss": 0.3932,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.1025184914469719,
+      "step": 5,
+      "valid_targets_mean": 14021.2,
+      "valid_targets_min": 7176
+    },
+    {
+      "epoch": 2.6,
+      "grad_norm": 0.2512775401674484,
+      "learning_rate": 3.457937254842823e-05,
+      "loss": 0.2862,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.08371572941541672,
+      "step": 10,
+      "valid_targets_mean": 14427.7,
+      "valid_targets_min": 2828
+    },
+    {
+      "epoch": 3.9,
+      "grad_norm": 0.1794639065475541,
+      "learning_rate": 2.37476262917145e-05,
+      "loss": 0.2367,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.07484349608421326,
+      "step": 15,
+      "valid_targets_mean": 13976.7,
+      "valid_targets_min": 9420
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.20840397498312163,
+      "learning_rate": 1.1484414168698547e-05,
+      "loss": 0.2056,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.19924341142177582,
+      "step": 20,
+      "valid_targets_mean": 14033.0,
+      "valid_targets_min": 5856
+    },
+    {
+      "epoch": 6.3,
+      "grad_norm": 0.12391804916466334,
+      "learning_rate": 2.473866399122733e-06,
+      "loss": 0.1917,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.062275081872940063,
+      "step": 25,
+      "valid_targets_mean": 14145.2,
+      "valid_targets_min": 8865
+    },
+    {
+      "epoch": 7.0,
+      "step": 28,
+      "total_flos": 1.6389573619325338e+17,
+      "train_loss": 0.0,
+      "train_runtime": 0.905,
+      "train_samples_per_second": 2444.322,
+      "train_steps_per_second": 30.941
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 28,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.6389573619325338e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}