初始化项目，由ModelHub XC社区提供模型

Model: thangvip/vwen-0.5 Source: Original Platform
2026-04-23 12:21:09 +08:00
commit 746289923f
85 changed files with 2728578 additions and 0 deletions
--- a/checkpoint-7568/trainer_state.json
+++ b/checkpoint-7568/trainer_state.json
@@ -0,0 +1,126 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.1047078223889412,
+  "eval_steps": 500,
+  "global_step": 7568,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "grad_norm": 1.4067256450653076,
+      "learning_rate": 9.930821699849192e-05,
+      "loss": 2.2227,
+      "step": 500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 2.9092910289764404,
+      "learning_rate": 9.861643399698382e-05,
+      "loss": 2.1156,
+      "step": 1000
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 1.1342206001281738,
+      "learning_rate": 9.792465099547574e-05,
+      "loss": 2.0774,
+      "step": 1500
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 3.928356409072876,
+      "learning_rate": 9.723286799396766e-05,
+      "loss": 2.0468,
+      "step": 2000
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.9985896944999695,
+      "learning_rate": 9.654108499245957e-05,
+      "loss": 2.0288,
+      "step": 2500
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 1.6732710599899292,
+      "learning_rate": 9.584930199095148e-05,
+      "loss": 1.9954,
+      "step": 3000
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.9810203313827515,
+      "learning_rate": 9.515751898944339e-05,
+      "loss": 1.9757,
+      "step": 3500
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.8682870864868164,
+      "learning_rate": 9.446573598793531e-05,
+      "loss": 1.9651,
+      "step": 4000
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.9405160546302795,
+      "learning_rate": 9.377395298642722e-05,
+      "loss": 1.9498,
+      "step": 4500
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.8670147061347961,
+      "learning_rate": 9.308216998491913e-05,
+      "loss": 1.935,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.9158061146736145,
+      "learning_rate": 9.239038698341104e-05,
+      "loss": 1.9252,
+      "step": 5500
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.8484827280044556,
+      "learning_rate": 9.169860398190296e-05,
+      "loss": 1.9001,
+      "step": 6000
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.8079173564910889,
+      "learning_rate": 9.100682098039488e-05,
+      "loss": 1.9174,
+      "step": 6500
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.8451229929924011,
+      "learning_rate": 9.031503797888679e-05,
+      "loss": 1.8939,
+      "step": 7000
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.7751355171203613,
+      "learning_rate": 8.962325497737871e-05,
+      "loss": 1.8915,
+      "step": 7500
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 72277,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 16,
+  "total_flos": 6.903809319402209e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}