初始化项目，由ModelHub XC社区提供模型

Model: lldois/v44_v29_clean_live_temporal_lr6e7_ep045
Source: Original Platform

trainer_state.json

@@ -0,0 +1,71 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.4584527220630373,
+  "eval_steps": 500,
+  "global_step": 40,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11461318051575932,
+      "grad_norm": 7.6875,
+      "learning_rate": 5.511499434787585e-07,
+      "loss": 2.213998794555664,
+      "step": 10
+    },
+    {
+      "epoch": 0.22922636103151864,
+      "grad_norm": 6.5625,
+      "learning_rate": 3.493783770842202e-07,
+      "loss": 2.0011642456054686,
+      "step": 20
+    },
+    {
+      "epoch": 0.3438395415472779,
+      "grad_norm": 5.6875,
+      "learning_rate": 1.1573618619309965e-07,
+      "loss": 1.9670379638671875,
+      "step": 30
+    },
+    {
+      "epoch": 0.4584527220630373,
+      "grad_norm": 6.40625,
+      "learning_rate": 1.0246520979990458e-09,
+      "loss": 1.946934700012207,
+      "step": 40
+    },
+    {
+      "epoch": 0.4584527220630373,
+      "step": 40,
+      "total_flos": 1.8995601921736704e+16,
+      "train_loss": 2.0322839260101317,
+      "train_runtime": 332.2523,
+      "train_samples_per_second": 0.473,
+      "train_steps_per_second": 0.12
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 40,
+  "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": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.8995601921736704e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}