初始化项目，由ModelHub XC社区提供模型

Model: cs-552-2026-MMRF/general_knowledge_model Source: Original Platform
2026-06-13 09:58:16 +08:00
commit 7c4ed61457
152 changed files with 12557 additions and 0 deletions
--- a/checkpoint-500/trainer_state.json
+++ b/checkpoint-500/trainer_state.json
@@ -0,0 +1,134 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.08012499499218781,
+  "eval_steps": 500,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.223189059495926,
+      "epoch": 0.008012499499218782,
+      "grad_norm": 2.875,
+      "learning_rate": 1.9842973882390643e-05,
+      "loss": 2.3428109741210936,
+      "mean_token_accuracy": 0.4963963569700718,
+      "num_tokens": 305366.0,
+      "step": 50
+    },
+    {
+      "entropy": 2.188130117058754,
+      "epoch": 0.016024998998437564,
+      "grad_norm": 3.328125,
+      "learning_rate": 1.96827431501362e-05,
+      "loss": 2.225358734130859,
+      "mean_token_accuracy": 0.5110921229422093,
+      "num_tokens": 605042.0,
+      "step": 100
+    },
+    {
+      "entropy": 2.197347580194473,
+      "epoch": 0.024037498497656344,
+      "grad_norm": 3.453125,
+      "learning_rate": 1.952251241788175e-05,
+      "loss": 2.2299420166015627,
+      "mean_token_accuracy": 0.5093490639328957,
+      "num_tokens": 906888.0,
+      "step": 150
+    },
+    {
+      "entropy": 2.1913950395584108,
+      "epoch": 0.03204999799687513,
+      "grad_norm": 2.90625,
+      "learning_rate": 1.9362281685627303e-05,
+      "loss": 2.210995941162109,
+      "mean_token_accuracy": 0.5097909501194954,
+      "num_tokens": 1213125.0,
+      "step": 200
+    },
+    {
+      "entropy": 2.1611998677253723,
+      "epoch": 0.040062497496093905,
+      "grad_norm": 3.25,
+      "learning_rate": 1.920205095337286e-05,
+      "loss": 2.1867474365234374,
+      "mean_token_accuracy": 0.5154410660266876,
+      "num_tokens": 1509026.0,
+      "step": 250
+    },
+    {
+      "entropy": 2.168051454424858,
+      "epoch": 0.04807499699531269,
+      "grad_norm": 2.9375,
+      "learning_rate": 1.904182022111841e-05,
+      "loss": 2.190071716308594,
+      "mean_token_accuracy": 0.5156884534657001,
+      "num_tokens": 1806850.0,
+      "step": 300
+    },
+    {
+      "entropy": 2.1737154543399813,
+      "epoch": 0.05608749649453147,
+      "grad_norm": 3.421875,
+      "learning_rate": 1.8881589488863967e-05,
+      "loss": 2.199147186279297,
+      "mean_token_accuracy": 0.514045044630766,
+      "num_tokens": 2104505.0,
+      "step": 350
+    },
+    {
+      "entropy": 2.148039159178734,
+      "epoch": 0.06409999599375026,
+      "grad_norm": 3.109375,
+      "learning_rate": 1.872135875660952e-05,
+      "loss": 2.176946105957031,
+      "mean_token_accuracy": 0.5177112428843975,
+      "num_tokens": 2391088.0,
+      "step": 400
+    },
+    {
+      "entropy": 2.194890711903572,
+      "epoch": 0.07211249549296903,
+      "grad_norm": 3.28125,
+      "learning_rate": 1.856112802435507e-05,
+      "loss": 2.216288757324219,
+      "mean_token_accuracy": 0.5110335703194141,
+      "num_tokens": 2691272.0,
+      "step": 450
+    },
+    {
+      "entropy": 2.1557786524295808,
+      "epoch": 0.08012499499218781,
+      "grad_norm": 4.15625,
+      "learning_rate": 1.8400897292100627e-05,
+      "loss": 2.1812680053710936,
+      "mean_token_accuracy": 0.5176697050035,
+      "num_tokens": 2991855.0,
+      "step": 500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 6241,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "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": 3.560511422243635e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}