初始化项目，由ModelHub XC社区提供模型

Model: princeton-nlp/Sheared-LLaMA-1.3B-ShareGPT Source: Original Platform
2026-05-18 11:19:02 +08:00
commit 09524b7992
13 changed files with 313 additions and 0 deletions
--- a/trainer_state.json
+++ b/trainer_state.json
@@ -0,0 +1,123 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 158,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.986836074908616e-05,
+      "loss": 1.2069,
+      "step": 10
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.882355001067892e-05,
+      "loss": 1.1431,
+      "step": 20
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.677782485216644e-05,
+      "loss": 1.1003,
+      "step": 30
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.381713366536311e-05,
+      "loss": 1.1053,
+      "step": 40
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.0065865909481417e-05,
+      "loss": 1.0894,
+      "step": 50
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.568162605525953e-05,
+      "loss": 1.1011,
+      "step": 60
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.084861204504122e-05,
+      "loss": 1.0828,
+      "step": 70
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5769876463904265e-05,
+      "loss": 1.0248,
+      "step": 80
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.0658795558326743e-05,
+      "loss": 0.6882,
+      "step": 90
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.5730104520100982e-05,
+      "loss": 0.6468,
+      "step": 100
+    },
+    {
+      "epoch": 1.27,
+      "eval_loss": 1.0625633001327515,
+      "eval_runtime": 0.5552,
+      "eval_samples_per_second": 18.01,
+      "eval_steps_per_second": 1.801,
+      "step": 100
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1190875675987356e-05,
+      "loss": 0.6445,
+      "step": 110
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 7.231818622338823e-06,
+      "loss": 0.6496,
+      "step": 120
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 4.019267817841835e-06,
+      "loss": 0.6305,
+      "step": 130
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.6881942648911076e-06,
+      "loss": 0.6451,
+      "step": 140
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.3653488440851255e-07,
+      "loss": 0.6417,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "step": 158,
+      "total_flos": 8944339187138560.0,
+      "train_loss": 0.8824256341668624,
+      "train_runtime": 246.3648,
+      "train_samples_per_second": 81.099,
+      "train_steps_per_second": 0.641
+    }
+  ],
+  "max_steps": 158,
+  "num_train_epochs": 2,
+  "total_flos": 8944339187138560.0,
+  "trial_name": null,
+  "trial_params": null
+}