Qwen2.5-1.5B-Open-R1-Distill / train_results.json
rainyuniverse's picture
Model save
9d68a25 verified
raw
history blame
214 Bytes
{
"total_flos": 488578517827584.0,
"train_loss": 0.5429407808540473,
"train_runtime": 24734.1783,
"train_samples": 93733,
"train_samples_per_second": 1.387,
"train_steps_per_second": 0.087
}