Qwen2.5-1.5B-Open-R1-Distill / train_results.json
MatchaLwc's picture
Model save
ea5f660 verified
{
"total_flos": 488621249396736.0,
"train_loss": 0.5498761394019447,
"train_runtime": 14566.6023,
"train_samples": 93733,
"train_samples_per_second": 2.355,
"train_steps_per_second": 0.074
}