初始化项目，由ModelHub XC社区提供模型

Model: laion/allenai-sera-unified-316__Qwen3-8B Source: Original Platform
2026-05-07 10:46:48 +08:00
commit 8b164a2a86
23 changed files with 152531 additions and 0 deletions
--- a/trainer_state.json
+++ b/trainer_state.json
@@ -0,0 +1,102 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 28,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.3,
+      "grad_norm": 2.0626110189657902,
+      "learning_rate": 3.9842294026289565e-05,
+      "loss": 0.4037,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.11264239251613617,
+      "step": 5,
+      "valid_targets_mean": 14021.2,
+      "valid_targets_min": 7176
+    },
+    {
+      "epoch": 2.6,
+      "grad_norm": 0.39234429982005087,
+      "learning_rate": 3.457937254842823e-05,
+      "loss": 0.3317,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.1028408631682396,
+      "step": 10,
+      "valid_targets_mean": 14427.7,
+      "valid_targets_min": 2828
+    },
+    {
+      "epoch": 3.9,
+      "grad_norm": 0.2713369990347595,
+      "learning_rate": 2.37476262917145e-05,
+      "loss": 0.3036,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.09833219647407532,
+      "step": 15,
+      "valid_targets_mean": 13976.7,
+      "valid_targets_min": 9420
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.2499268308207944,
+      "learning_rate": 1.1484414168698547e-05,
+      "loss": 0.2798,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.2743203639984131,
+      "step": 20,
+      "valid_targets_mean": 14033.0,
+      "valid_targets_min": 5856
+    },
+    {
+      "epoch": 6.3,
+      "grad_norm": 0.1760972086121257,
+      "learning_rate": 2.473866399122733e-06,
+      "loss": 0.2712,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.08980667591094971,
+      "step": 25,
+      "valid_targets_mean": 14145.2,
+      "valid_targets_min": 8865
+    },
+    {
+      "epoch": 7.0,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.2726576030254364,
+      "step": 28,
+      "total_flos": 1.6389573619325338e+17,
+      "train_loss": 0.3133217862674168,
+      "train_runtime": 739.5307,
+      "train_samples_per_second": 2.991,
+      "train_steps_per_second": 0.038,
+      "valid_targets_mean": 13154.9,
+      "valid_targets_min": 5475
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 28,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 300,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.6389573619325338e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}