初始化项目，由ModelHub XC社区提供模型

Model: NastasiaM/mbert-loraxs-qa-vanilla Source: Original Platform
2026-06-10 00:18:47 +08:00
commit d696db249d
795 changed files with 8285474 additions and 0 deletions
--- a/run-8/checkpoint-3170/trainer_state.json
+++ b/run-8/checkpoint-3170/trainer_state.json
@@ -0,0 +1,126 @@
+{
+  "best_global_step": 2536,
+  "best_metric": 75.87754861144096,
+  "best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-8/checkpoint-2536",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 3170,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 13.794865608215332,
+      "learning_rate": 0.00014003511035757225,
+      "loss": 3.036647086639886,
+      "step": 634
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 55.76,
+      "eval_f1": 69.6160406135031,
+      "eval_loss": 1.4716368913650513,
+      "eval_runtime": 7.8535,
+      "eval_samples_per_second": 161.074,
+      "eval_steps_per_second": 10.187,
+      "step": 634
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 9.303217887878418,
+      "learning_rate": 0.00010504013204984627,
+      "loss": 1.4382544676964215,
+      "step": 1268
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 58.32,
+      "eval_f1": 72.97295260861294,
+      "eval_loss": 1.3276563882827759,
+      "eval_runtime": 7.7377,
+      "eval_samples_per_second": 163.485,
+      "eval_steps_per_second": 10.339,
+      "step": 1268
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 5.403870105743408,
+      "learning_rate": 7.00451537421203e-05,
+      "loss": 1.1502308168621846,
+      "step": 1902
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 60.72,
+      "eval_f1": 74.80134341907049,
+      "eval_loss": 1.259326457977295,
+      "eval_runtime": 7.8558,
+      "eval_samples_per_second": 161.027,
+      "eval_steps_per_second": 10.184,
+      "step": 1902
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 9.722108840942383,
+      "learning_rate": 3.505017543439432e-05,
+      "loss": 0.9732898988934345,
+      "step": 2536
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 61.44,
+      "eval_f1": 75.87754861144096,
+      "eval_loss": 1.28603994846344,
+      "eval_runtime": 7.8253,
+      "eval_samples_per_second": 161.655,
+      "eval_steps_per_second": 10.223,
+      "step": 2536
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 11.066482543945312,
+      "learning_rate": 5.519712666833751e-08,
+      "loss": 0.8625883950796037,
+      "step": 3170
+    },
+    {
+      "epoch": 5.0,
+      "eval_exact_match": 60.64,
+      "eval_f1": 75.23748723190614,
+      "eval_loss": 1.321590542793274,
+      "eval_runtime": 7.8071,
+      "eval_samples_per_second": 162.033,
+      "eval_steps_per_second": 10.247,
+      "step": 3170
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 3170,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "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": 1.545247865581056e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 0.00015747740238476691,
+    "lora_alpha": 128,
+    "lora_dropout": 0.10139328835291189,
+    "lora_r": 256,
+    "num_train_epochs": 5,
+    "weight_decay": 0.05287944657262306
+  }
+}