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-5/checkpoint-3170/trainer_state.json
+++ b/run-5/checkpoint-3170/trainer_state.json
@@ -0,0 +1,126 @@
+{
+  "best_global_step": 3170,
+  "best_metric": 66.21587013990109,
+  "best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-5/checkpoint-3170",
+  "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": 6.493763446807861,
+      "learning_rate": 0.00013719632775176179,
+      "loss": 4.3875401253203865,
+      "step": 634
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 32.96,
+      "eval_f1": 44.88224715216386,
+      "eval_loss": 2.3734943866729736,
+      "eval_runtime": 7.7543,
+      "eval_samples_per_second": 163.136,
+      "eval_steps_per_second": 10.317,
+      "step": 634
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 5.867770671844482,
+      "learning_rate": 0.00010291076535735226,
+      "loss": 2.3487211883256114,
+      "step": 1268
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 46.56,
+      "eval_f1": 59.14511314857039,
+      "eval_loss": 1.8806413412094116,
+      "eval_runtime": 7.6953,
+      "eval_samples_per_second": 164.386,
+      "eval_steps_per_second": 10.396,
+      "step": 1268
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 5.569732189178467,
+      "learning_rate": 6.862520296294273e-05,
+      "loss": 2.0055035612184544,
+      "step": 1902
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 49.92,
+      "eval_f1": 63.30056490097892,
+      "eval_loss": 1.6955205202102661,
+      "eval_runtime": 7.6628,
+      "eval_samples_per_second": 165.083,
+      "eval_steps_per_second": 10.44,
+      "step": 1902
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 5.1621527671813965,
+      "learning_rate": 3.43396405685332e-05,
+      "loss": 1.855263502439866,
+      "step": 2536
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 52.56,
+      "eval_f1": 65.78508480758633,
+      "eval_loss": 1.6228810548782349,
+      "eval_runtime": 7.665,
+      "eval_samples_per_second": 165.037,
+      "eval_steps_per_second": 10.437,
+      "step": 2536
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 6.866057872772217,
+      "learning_rate": 5.407817412367433e-08,
+      "loss": 1.7964960814274449,
+      "step": 3170
+    },
+    {
+      "epoch": 5.0,
+      "eval_exact_match": 52.8,
+      "eval_f1": 66.21587013990109,
+      "eval_loss": 1.611862301826477,
+      "eval_runtime": 7.7323,
+      "eval_samples_per_second": 163.599,
+      "eval_steps_per_second": 10.346,
+      "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": 1.12139369540352e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 0.00015428503077484287,
+    "lora_alpha": 32,
+    "lora_dropout": 0.12243934543152664,
+    "lora_r": 64,
+    "num_train_epochs": 5,
+    "weight_decay": 0.021835569485970197
+  }
+}