初始化项目，由ModelHub XC社区提供模型

Model: uyenlk/UNDIAL_forget10_5e-5_Llama-3.2-3B-Instruct_alpha1_beta30 Source: Original Platform
2026-05-12 18:36:22 +08:00
commit 03e234caad
19 changed files with 41763 additions and 0 deletions
--- a/trainer_state.json
+++ b/trainer_state.json
@@ -0,0 +1,126 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.8,
+  "eval_steps": 500,
+  "global_step": 60,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.4,
+      "grad_norm": 25.375,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 4.3411,
+      "step": 5
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 10.875,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 3.56,
+      "step": 10
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 9.0625,
+      "learning_rate": 4.6875e-05,
+      "loss": 3.1256,
+      "step": 15
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 9.125,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 2.8624,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 7.9375,
+      "learning_rate": 3.6458333333333336e-05,
+      "loss": 2.7787,
+      "step": 25
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 6.59375,
+      "learning_rate": 3.125e-05,
+      "loss": 2.5525,
+      "step": 30
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 6.625,
+      "learning_rate": 2.604166666666667e-05,
+      "loss": 2.5085,
+      "step": 35
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 9.125,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 2.4691,
+      "step": 40
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 7.28125,
+      "learning_rate": 1.5625e-05,
+      "loss": 2.3675,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 5.96875,
+      "learning_rate": 1.0416666666666668e-05,
+      "loss": 2.3596,
+      "step": 50
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 6.96875,
+      "learning_rate": 5.208333333333334e-06,
+      "loss": 2.3325,
+      "step": 55
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 5.28125,
+      "learning_rate": 0.0,
+      "loss": 2.3121,
+      "step": 60
+    },
+    {
+      "epoch": 4.8,
+      "step": 60,
+      "total_flos": 0.0,
+      "train_loss": 2.7974623521169026,
+      "train_runtime": 818.9975,
+      "train_samples_per_second": 2.442,
+      "train_steps_per_second": 0.073
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 60,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}