Files
qwen3-4b-instruct-2507-full…/train_results.json
ModelHub XC fe673a1dc2 初始化项目,由ModelHub XC社区提供模型
Model: shubhamrgandhi/qwen3-4b-instruct-2507-full-sft-prm-r2egym-swebench-instructions-k5-cwm-plus-qwen
Source: Original Platform
2026-08-22 20:13:16 +08:00

8 lines
206 B
JSON

{
"epoch": 3.0,
"total_flos": 1.329254778357678e+18,
"train_loss": 0.361527387419449,
"train_runtime": 121252.8356,
"train_samples_per_second": 0.16,
"train_steps_per_second": 0.02
}