初始化项目,由ModelHub XC社区提供模型

Model: jackf857/qwen3-8b-base-epsilon-dpo-hh-harmless-4xh200-batch-64-20260424-040415
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-14 00:59:58 +08:00
commit 42e9e465a7
23 changed files with 168074 additions and 0 deletions

9
train_results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"epoch": 0.999244142101285,
"total_flos": 0.0,
"train_loss": 1.1575256359739492,
"train_runtime": 3253.0214,
"train_samples": 42336,
"train_samples_per_second": 13.014,
"train_steps_per_second": 0.203
}