初始化项目，由ModelHub XC社区提供模型

Model: laion/nemotron-terminal-corpus-unified-316__Qwen3-8B Source: Original Platform
2026-05-27 23:20:23 +08:00
commit 85390b86ad
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": 1.7363245597604395,
+      "learning_rate": 3.9842294026289565e-05,
+      "loss": 0.8962,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.3306814432144165,
+      "step": 5,
+      "valid_targets_mean": 12824.2,
+      "valid_targets_min": 1516
+    },
+    {
+      "epoch": 2.6,
+      "grad_norm": 1.0398281811830472,
+      "learning_rate": 3.457937254842823e-05,
+      "loss": 0.7514,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.20077958703041077,
+      "step": 10,
+      "valid_targets_mean": 10740.1,
+      "valid_targets_min": 2488
+    },
+    {
+      "epoch": 3.9,
+      "grad_norm": 0.6180460706018052,
+      "learning_rate": 2.37476262917145e-05,
+      "loss": 0.7133,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.27195170521736145,
+      "step": 15,
+      "valid_targets_mean": 12384.0,
+      "valid_targets_min": 1760
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.5352399487217326,
+      "learning_rate": 1.1484414168698547e-05,
+      "loss": 0.6714,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.6289256811141968,
+      "step": 20,
+      "valid_targets_mean": 9918.0,
+      "valid_targets_min": 913
+    },
+    {
+      "epoch": 6.3,
+      "grad_norm": 0.3384143471156126,
+      "learning_rate": 2.473866399122733e-06,
+      "loss": 0.6624,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.21406814455986023,
+      "step": 25,
+      "valid_targets_mean": 10884.2,
+      "valid_targets_min": 1760
+    },
+    {
+      "epoch": 7.0,
+      "loss_nan_ranks": 0,
+      "loss_rank_avg": 0.6334422826766968,
+      "step": 28,
+      "total_flos": 8.133461226500915e+16,
+      "train_loss": 0.7130181619099208,
+      "train_runtime": 707.1363,
+      "train_samples_per_second": 3.128,
+      "train_steps_per_second": 0.04,
+      "valid_targets_mean": 12522.3,
+      "valid_targets_min": 2429
+    }
+  ],
+  "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": 8.133461226500915e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}