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-2/checkpoint-2536/trainer_state.json
+++ b/run-2/checkpoint-2536/trainer_state.json
@@ -0,0 +1,109 @@
+{
+  "best_global_step": 2536,
+  "best_metric": 62.864863739540695,
+  "best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-2/checkpoint-2536",
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 2536,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 12.638736724853516,
+      "learning_rate": 0.0004141217931379858,
+      "loss": 3.862071329110804,
+      "step": 634
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 38.32,
+      "eval_f1": 50.68513380792723,
+      "eval_loss": 2.2031736373901367,
+      "eval_runtime": 7.6317,
+      "eval_samples_per_second": 165.756,
+      "eval_steps_per_second": 10.483,
+      "step": 634
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 9.9076566696167,
+      "learning_rate": 0.0002761537338371539,
+      "loss": 2.3274724821939077,
+      "step": 1268
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 45.2,
+      "eval_f1": 57.96122541995168,
+      "eval_loss": 1.8692753314971924,
+      "eval_runtime": 7.5957,
+      "eval_samples_per_second": 166.542,
+      "eval_steps_per_second": 10.532,
+      "step": 1268
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 9.15977954864502,
+      "learning_rate": 0.00013818567453632212,
+      "loss": 2.07331573775138,
+      "step": 1902
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 48.48,
+      "eval_f1": 61.9201483756793,
+      "eval_loss": 1.7289328575134277,
+      "eval_runtime": 7.6409,
+      "eval_samples_per_second": 165.556,
+      "eval_steps_per_second": 10.47,
+      "step": 1902
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 9.809730529785156,
+      "learning_rate": 2.1761523549027105e-07,
+      "loss": 1.9579327219292193,
+      "step": 2536
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 49.6,
+      "eval_f1": 62.864863739540695,
+      "eval_loss": 1.7020000219345093,
+      "eval_runtime": 7.6439,
+      "eval_samples_per_second": 165.492,
+      "eval_steps_per_second": 10.466,
+      "step": 2536
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2536,
+  "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": 8200975522295808.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 0.0004965979673887985,
+    "lora_alpha": 32,
+    "lora_dropout": 0.009962720741248932,
+    "lora_r": 16,
+    "num_train_epochs": 4,
+    "weight_decay": 0.06397281536936893
+  }
+}