初始化项目，由ModelHub XC社区提供模型

Model: locuslab/tofu_ft_llama2-7b Source: Original Platform
2026-07-12 11:56:12 +08:00
commit d749ad3ccc
15 changed files with 94100 additions and 0 deletions
--- a/trainer_state.json
+++ b/trainer_state.json
@@ -0,0 +1,141 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 625,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.25,
+      "learning_rate": 5e-06,
+      "loss": 2.1158,
+      "step": 31
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.3913,
+      "step": 62
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.449378330373002e-06,
+      "loss": 1.2838,
+      "step": 93
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 8.898756660746004e-06,
+      "loss": 1.2016,
+      "step": 124
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 8.348134991119005e-06,
+      "loss": 1.0974,
+      "step": 155
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.797513321492007e-06,
+      "loss": 0.8999,
+      "step": 186
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.246891651865009e-06,
+      "loss": 0.7662,
+      "step": 217
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 6.6962699822380115e-06,
+      "loss": 0.7183,
+      "step": 248
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 6.145648312611013e-06,
+      "loss": 0.6274,
+      "step": 279
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 5.595026642984015e-06,
+      "loss": 0.4377,
+      "step": 310
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 5.044404973357016e-06,
+      "loss": 0.2646,
+      "step": 341
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.493783303730018e-06,
+      "loss": 0.2512,
+      "step": 372
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 3.94316163410302e-06,
+      "loss": 0.2128,
+      "step": 403
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 3.3925399644760213e-06,
+      "loss": 0.1259,
+      "step": 434
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 2.841918294849023e-06,
+      "loss": 0.0753,
+      "step": 465
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 2.291296625222025e-06,
+      "loss": 0.0686,
+      "step": 496
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 1.7406749555950267e-06,
+      "loss": 0.054,
+      "step": 527
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 1.1900532859680285e-06,
+      "loss": 0.0332,
+      "step": 558
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 6.394316163410303e-07,
+      "loss": 0.0204,
+      "step": 589
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 8.880994671403197e-08,
+      "loss": 0.0144,
+      "step": 620
+    }
+  ],
+  "logging_steps": 31,
+  "max_steps": 625,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 625,
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}