初始化项目，由ModelHub XC社区提供模型

Model: dwaru/RuAdaptGPT2Small
Source: Original Platform

trainer_state.json

@@ -0,0 +1,105 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 2388,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 1.082448959350586,
+      "learning_rate": 3.953098827470687e-05,
+      "loss": 1.6167,
+      "step": 500
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "eval_CEFR_f1": 0.3644590403211093,
+      "eval_loss": 1.5046868324279785,
+      "eval_rouge1": 0.31,
+      "eval_rouge2": 0.13131313131313133,
+      "eval_runtime": 15.7708,
+      "eval_samples_per_second": 6.341,
+      "eval_steps_per_second": 0.127,
+      "step": 500
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 1.0536589622497559,
+      "learning_rate": 2.906197654941374e-05,
+      "loss": 1.3693,
+      "step": 1000
+    },
+    {
+      "epoch": 1.675041876046901,
+      "eval_CEFR_f1": 0.3023575040876079,
+      "eval_loss": 1.43952214717865,
+      "eval_rouge1": 0.35,
+      "eval_rouge2": 0.15151515151515152,
+      "eval_runtime": 17.1613,
+      "eval_samples_per_second": 5.827,
+      "eval_steps_per_second": 0.117,
+      "step": 1000
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 1.0843034982681274,
+      "learning_rate": 1.8592964824120602e-05,
+      "loss": 1.2575,
+      "step": 1500
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "eval_CEFR_f1": 0.30705782312925173,
+      "eval_loss": 1.4209942817687988,
+      "eval_rouge1": 0.33,
+      "eval_rouge2": 0.13131313131313133,
+      "eval_runtime": 16.7011,
+      "eval_samples_per_second": 5.988,
+      "eval_steps_per_second": 0.12,
+      "step": 1500
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "grad_norm": 1.1220632791519165,
+      "learning_rate": 8.123953098827471e-06,
+      "loss": 1.1912,
+      "step": 2000
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "eval_CEFR_f1": 0.31766408627240766,
+      "eval_loss": 1.4111934900283813,
+      "eval_rouge1": 0.34,
+      "eval_rouge2": 0.1414141414141414,
+      "eval_runtime": 15.8447,
+      "eval_samples_per_second": 6.311,
+      "eval_steps_per_second": 0.126,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2388,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "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": 1.2855167870976e+16,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}