Qwen-2.5-3B-Simple-RL / train_results.json
JeffP111's picture
Model save
f087f56 verified
{
"total_flos": 0.0,
"train_loss": 25.241294363718474,
"train_runtime": 70314.3192,
"train_samples": 7500,
"train_samples_per_second": 0.32,
"train_steps_per_second": 0.003
}