Qwen2.5-1.5B-Open-R1-Distill / all_results.json
tyfeng1997's picture
End of training
1fdf373 verified
raw
history blame contribute delete
382 Bytes
{
"eval_loss": 0.8344451189041138,
"eval_runtime": 9.7847,
"eval_samples": 100,
"eval_samples_per_second": 13.184,
"eval_steps_per_second": 3.373,
"total_flos": 6.955833048956928e+17,
"train_loss": 0.8332233665607593,
"train_runtime": 5910.2876,
"train_samples": 16610,
"train_samples_per_second": 3.659,
"train_steps_per_second": 0.114
}