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-3/checkpoint-3170/trainer_state.json
+++ b/run-3/checkpoint-3170/trainer_state.json
@@ -0,0 +1,126 @@
+{
+  "best_global_step": 2536,
+  "best_metric": 73.9060878995219,
+  "best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-3/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": 25.72856330871582,
+      "learning_rate": 0.00036170225367171497,
+      "loss": 2.8233307558655363,
+      "step": 634
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 50.32,
+      "eval_f1": 63.941961174475914,
+      "eval_loss": 1.6252752542495728,
+      "eval_runtime": 7.6169,
+      "eval_samples_per_second": 166.078,
+      "eval_steps_per_second": 10.503,
+      "step": 634
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 27.754880905151367,
+      "learning_rate": 0.0002893846160722614,
+      "loss": 1.7085243971189865,
+      "step": 1268
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 55.28,
+      "eval_f1": 70.42089105490078,
+      "eval_loss": 1.4321246147155762,
+      "eval_runtime": 7.709,
+      "eval_samples_per_second": 164.095,
+      "eval_steps_per_second": 10.378,
+      "step": 1268
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 28.5296688079834,
+      "learning_rate": 0.00021706697847280782,
+      "loss": 1.4246620154155165,
+      "step": 1902
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 56.88,
+      "eval_f1": 71.30211904005495,
+      "eval_loss": 1.389769434928894,
+      "eval_runtime": 7.7582,
+      "eval_samples_per_second": 163.053,
+      "eval_steps_per_second": 10.312,
+      "step": 1902
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 20.22711944580078,
+      "learning_rate": 0.00014474934087335426,
+      "loss": 1.1686882717000198,
+      "step": 2536
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 59.36,
+      "eval_f1": 73.9060878995219,
+      "eval_loss": 1.3756252527236938,
+      "eval_runtime": 7.8225,
+      "eval_samples_per_second": 161.714,
+      "eval_steps_per_second": 10.227,
+      "step": 2536
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 13.985761642456055,
+      "learning_rate": 7.243170327390066e-05,
+      "loss": 0.9745011675621057,
+      "step": 3170
+    },
+    {
+      "epoch": 5.0,
+      "eval_exact_match": 59.68,
+      "eval_f1": 73.67245039400858,
+      "eval_loss": 1.39497971534729,
+      "eval_runtime": 7.6711,
+      "eval_samples_per_second": 164.904,
+      "eval_steps_per_second": 10.429,
+      "step": 3170
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 3804,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.255786481069568e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 0.0003904468036323811,
+    "lora_alpha": 256,
+    "lora_dropout": 0.22980730880873645,
+    "lora_r": 128,
+    "num_train_epochs": 6,
+    "weight_decay": 0.06293800360990247
+  }
+}