初始化项目，由ModelHub XC社区提供模型

Model: lldois/v24_v15_cot_light_repair_lr25e6_ep035 Source: Original Platform
2026-07-18 07:37:10 +08:00
commit 3e0527763f
17 changed files with 2734 additions and 0 deletions
--- a/trainer_state.json
+++ b/trainer_state.json
@@ -0,0 +1,190 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.351575456053068,
+  "eval_steps": 500,
+  "global_step": 106,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01658374792703151,
+      "grad_norm": 26.625,
+      "learning_rate": 2.4994186045291896e-06,
+      "loss": 1.9901918411254882,
+      "step": 5
+    },
+    {
+      "epoch": 0.03316749585406302,
+      "grad_norm": 21.75,
+      "learning_rate": 2.4791264942053156e-06,
+      "loss": 2.094209098815918,
+      "step": 10
+    },
+    {
+      "epoch": 0.04975124378109453,
+      "grad_norm": 19.25,
+      "learning_rate": 2.4303031863746234e-06,
+      "loss": 1.80328426361084,
+      "step": 15
+    },
+    {
+      "epoch": 0.06633499170812604,
+      "grad_norm": 26.375,
+      "learning_rate": 2.354081995993076e-06,
+      "loss": 2.002113914489746,
+      "step": 20
+    },
+    {
+      "epoch": 0.08291873963515754,
+      "grad_norm": 20.375,
+      "learning_rate": 2.2522322135667634e-06,
+      "loss": 1.559170913696289,
+      "step": 25
+    },
+    {
+      "epoch": 0.09950248756218906,
+      "grad_norm": 12.6875,
+      "learning_rate": 2.127118035353867e-06,
+      "loss": 1.7275423049926757,
+      "step": 30
+    },
+    {
+      "epoch": 0.11608623548922056,
+      "grad_norm": 23.125,
+      "learning_rate": 1.9816436842690527e-06,
+      "loss": 1.4820417404174804,
+      "step": 35
+    },
+    {
+      "epoch": 0.13266998341625208,
+      "grad_norm": 10.8125,
+      "learning_rate": 1.819185995375714e-06,
+      "loss": 1.4906288146972657,
+      "step": 40
+    },
+    {
+      "epoch": 0.14925373134328357,
+      "grad_norm": 12.6875,
+      "learning_rate": 1.6435160308307313e-06,
+      "loss": 1.7329635620117188,
+      "step": 45
+    },
+    {
+      "epoch": 0.16583747927031509,
+      "grad_norm": 36.75,
+      "learning_rate": 1.458711543801153e-06,
+      "loss": 1.8760557174682617,
+      "step": 50
+    },
+    {
+      "epoch": 0.1824212271973466,
+      "grad_norm": 10.5625,
+      "learning_rate": 1.2690623232911575e-06,
+      "loss": 1.546410369873047,
+      "step": 55
+    },
+    {
+      "epoch": 0.19900497512437812,
+      "grad_norm": 46.25,
+      "learning_rate": 1.0789706170701125e-06,
+      "loss": 1.8373964309692383,
+      "step": 60
+    },
+    {
+      "epoch": 0.2155887230514096,
+      "grad_norm": 54.25,
+      "learning_rate": 8.928489441425056e-07,
+      "loss": 2.1412117004394533,
+      "step": 65
+    },
+    {
+      "epoch": 0.23217247097844113,
+      "grad_norm": 17.75,
+      "learning_rate": 7.15017668796009e-07,
+      "loss": 1.5714065551757812,
+      "step": 70
+    },
+    {
+      "epoch": 0.24875621890547264,
+      "grad_norm": 10.8125,
+      "learning_rate": 5.496047137983128e-07,
+      "loss": 1.6369653701782227,
+      "step": 75
+    },
+    {
+      "epoch": 0.26533996683250416,
+      "grad_norm": 17.625,
+      "learning_rate": 4.0044974065819974e-07,
+      "loss": 1.8116127014160157,
+      "step": 80
+    },
+    {
+      "epoch": 0.28192371475953565,
+      "grad_norm": 12.1875,
+      "learning_rate": 2.7101502117425833e-07,
+      "loss": 1.5387956619262695,
+      "step": 85
+    },
+    {
+      "epoch": 0.29850746268656714,
+      "grad_norm": 12.3125,
+      "learning_rate": 1.6430506917256422e-07,
+      "loss": 1.850395393371582,
+      "step": 90
+    },
+    {
+      "epoch": 0.3150912106135987,
+      "grad_norm": 7.78125,
+      "learning_rate": 8.279689798808068e-08,
+      "loss": 1.588356113433838,
+      "step": 95
+    },
+    {
+      "epoch": 0.33167495854063017,
+      "grad_norm": 7.0,
+      "learning_rate": 2.8382522593593807e-08,
+      "loss": 1.513393497467041,
+      "step": 100
+    },
+    {
+      "epoch": 0.3482587064676617,
+      "grad_norm": 56.75,
+      "learning_rate": 2.3250410501327037e-09,
+      "loss": 1.58885498046875,
+      "step": 105
+    },
+    {
+      "epoch": 0.351575456053068,
+      "step": 106,
+      "total_flos": 5.062818542541619e+16,
+      "train_loss": 1.7334247456406646,
+      "train_runtime": 876.5295,
+      "train_samples_per_second": 0.482,
+      "train_steps_per_second": 0.121
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 106,
+  "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": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.062818542541619e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}