Qwen2.5-1.5B-Open-R1-Distill / train_results.json
yfliao's picture
Model save
1de8120 verified
raw
history blame contribute delete
213 Bytes
{
"total_flos": 488621249396736.0,
"train_loss": 0.579121366365632,
"train_runtime": 3078.0231,
"train_samples": 93733,
"train_samples_per_second": 11.144,
"train_steps_per_second": 0.087
}