Files
DeepSeek-R1-Distill-Qwen-7B…/all_results.json
ModelHub XC 9a7835e1f2 初始化项目,由ModelHub XC社区提供模型
Model: lightblue/DeepSeek-R1-Distill-Qwen-7B-Japanese
Source: Original Platform
2026-06-24 17:08:18 +08:00

12 lines
341 B
JSON

{
"epoch": 1.0,
"eval_loss": 0.4519386291503906,
"eval_runtime": 0.8439,
"eval_samples_per_second": 4.74,
"eval_steps_per_second": 1.185,
"total_flos": 7.03161217014825e+16,
"train_loss": 0.50457603516786,
"train_runtime": 362.0803,
"train_samples_per_second": 1.016,
"train_steps_per_second": 0.127
}