初始化项目，由ModelHub XC社区提供模型

Model: uyenlk/SimNPO_forget10_5e-5_Llama-3.2-3B-Instruct_gamma0.25_delta1.0_beta3.5 Source: Original Platform
2026-05-12 18:58:24 +08:00
commit 9935303364
19 changed files with 41773 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": 3.984375,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 0.5223,
+      "step": 5
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 16.5,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.5515,
+      "step": 10
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 13.875,
+      "learning_rate": 4.6875e-05,
+      "loss": 0.5093,
+      "step": 15
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 12.1875,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.5249,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 11.875,
+      "learning_rate": 3.6458333333333336e-05,
+      "loss": 0.4755,
+      "step": 25
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 8.375,
+      "learning_rate": 3.125e-05,
+      "loss": 0.356,
+      "step": 30
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 7.6875,
+      "learning_rate": 2.604166666666667e-05,
+      "loss": 0.3406,
+      "step": 35
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 13.5,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 0.3117,
+      "step": 40
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 6.34375,
+      "learning_rate": 1.5625e-05,
+      "loss": 0.2719,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 8.25,
+      "learning_rate": 1.0416666666666668e-05,
+      "loss": 0.2654,
+      "step": 50
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 6.90625,
+      "learning_rate": 5.208333333333334e-06,
+      "loss": 0.2696,
+      "step": 55
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 7.6875,
+      "learning_rate": 0.0,
+      "loss": 0.2359,
+      "step": 60
+    },
+    {
+      "epoch": 4.8,
+      "step": 60,
+      "total_flos": 0.0,
+      "train_loss": 0.3862247069676717,
+      "train_runtime": 1181.9694,
+      "train_samples_per_second": 1.692,
+      "train_steps_per_second": 0.051
+    }
+  ],
+  "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
+}