初始化项目，由ModelHub XC社区提供模型

Model: mohit-1710/loomstack-qwen-sft-terminal Source: Original Platform
2026-05-07 17:32:15 +08:00
commit bc45922a48
31 changed files with 1989 additions and 0 deletions
--- a/checkpoint-90/trainer_state.json
+++ b/checkpoint-90/trainer_state.json
@@ -0,0 +1,160 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 90,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 0.7361575961112976,
+      "learning_rate": 2e-05,
+      "loss": 0.7169363975524903,
+      "step": 5
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.32090041041374207,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6374710083007813,
+      "step": 10
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.5020814538002014,
+      "learning_rate": 4.75e-05,
+      "loss": 0.6157864570617676,
+      "step": 15
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3337880074977875,
+      "learning_rate": 4.4375e-05,
+      "loss": 0.5243042469024658,
+      "step": 20
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 0.523672342300415,
+      "learning_rate": 4.125e-05,
+      "loss": 0.45089244842529297,
+      "step": 25
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.38501089811325073,
+      "learning_rate": 3.8125e-05,
+      "loss": 0.41745529174804685,
+      "step": 30
+    },
+    {
+      "epoch": 1.1666666666666667,
+      "grad_norm": 0.333345502614975,
+      "learning_rate": 3.5e-05,
+      "loss": 0.3673891305923462,
+      "step": 35
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.28795114159584045,
+      "learning_rate": 3.1875e-05,
+      "loss": 0.3276315689086914,
+      "step": 40
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.4279637336730957,
+      "learning_rate": 2.8749999999999997e-05,
+      "loss": 0.2765533208847046,
+      "step": 45
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.32112613320350647,
+      "learning_rate": 2.5625e-05,
+      "loss": 0.2375420570373535,
+      "step": 50
+    },
+    {
+      "epoch": 1.8333333333333335,
+      "grad_norm": 0.36314302682876587,
+      "learning_rate": 2.25e-05,
+      "loss": 0.19293277263641356,
+      "step": 55
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.4400066137313843,
+      "learning_rate": 1.9375e-05,
+      "loss": 0.18874874114990234,
+      "step": 60
+    },
+    {
+      "epoch": 2.1666666666666665,
+      "grad_norm": 0.2499208152294159,
+      "learning_rate": 1.6250000000000002e-05,
+      "loss": 0.16871228218078613,
+      "step": 65
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.2517125606536865,
+      "learning_rate": 1.3125e-05,
+      "loss": 0.13855531215667724,
+      "step": 70
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.255470335483551,
+      "learning_rate": 1e-05,
+      "loss": 0.12582218647003174,
+      "step": 75
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.3331236243247986,
+      "learning_rate": 6.875000000000001e-06,
+      "loss": 0.09783701896667481,
+      "step": 80
+    },
+    {
+      "epoch": 2.8333333333333335,
+      "grad_norm": 0.33902686834335327,
+      "learning_rate": 3.75e-06,
+      "loss": 0.11323860883712769,
+      "step": 85
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.20473864674568176,
+      "learning_rate": 6.25e-07,
+      "loss": 0.11003848314285278,
+      "step": 90
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 90,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5444060958167040.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}