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-22/checkpoint-2536/trainer_state.json
+++ b/run-22/checkpoint-2536/trainer_state.json
@@ -0,0 +1,109 @@
+{
+  "best_global_step": 2536,
+  "best_metric": 75.90861914150922,
+  "best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-22/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": 6.846948146820068,
+      "learning_rate": 0.0002936503084341597,
+      "loss": 3.051067171788742,
+      "step": 634
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 55.92,
+      "eval_f1": 69.50774414487569,
+      "eval_loss": 1.4324498176574707,
+      "eval_runtime": 7.7695,
+      "eval_samples_per_second": 162.816,
+      "eval_steps_per_second": 10.297,
+      "step": 634
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 4.550962924957275,
+      "learning_rate": 0.0002349387677380836,
+      "loss": 1.4162167329517448,
+      "step": 1268
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 58.48,
+      "eval_f1": 72.4666290482864,
+      "eval_loss": 1.3376092910766602,
+      "eval_runtime": 7.8052,
+      "eval_samples_per_second": 162.072,
+      "eval_steps_per_second": 10.25,
+      "step": 1268
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.3896350860595703,
+      "learning_rate": 0.0001762272270420075,
+      "loss": 1.1300597566911474,
+      "step": 1902
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 61.44,
+      "eval_f1": 75.33434312606144,
+      "eval_loss": 1.2791812419891357,
+      "eval_runtime": 7.8106,
+      "eval_samples_per_second": 161.958,
+      "eval_steps_per_second": 10.242,
+      "step": 1902
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.916056156158447,
+      "learning_rate": 0.00011751568634593144,
+      "loss": 0.9393305733376873,
+      "step": 2536
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 62.48,
+      "eval_f1": 75.90861914150922,
+      "eval_loss": 1.2876638174057007,
+      "eval_runtime": 7.8355,
+      "eval_samples_per_second": 161.445,
+      "eval_steps_per_second": 10.21,
+      "step": 2536
+    }
+  ],
+  "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": 1.2361982924648448e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 0.0003169867567865432,
+    "lora_alpha": 64,
+    "lora_dropout": 0.07955181572910565,
+    "lora_r": 256,
+    "num_train_epochs": 6,
+    "weight_decay": 0.07964841643272833
+  }
+}