Files
OpenR1-Distill-Qwen3-8B-Med…/train_results.json

9 lines
238 B
JSON
Raw Permalink Normal View History

{
"epoch": 5.0,
"total_flos": 1.4678746016632013e+18,
"train_loss": 0.07142430102201106,
"train_runtime": 444789.0665,
"train_samples": 49704,
"train_samples_per_second": 0.559,
"train_steps_per_second": 0.14
}