Qwen2.5-1.5B-Open-R1-Distill / train_results.json
khalidalt's picture
Model save
da84ded verified
{
"total_flos": 76973799899136.0,
"train_loss": 0.802493240706314,
"train_runtime": 753.9282,
"train_samples": 16610,
"train_samples_per_second": 28.68,
"train_steps_per_second": 0.224
}