Qwen-2.5-7B-Simple-RL / train_results.json
yolay's picture
Model save
4315844 verified
{
"total_flos": 0.0,
"train_loss": 0.00286464851636153,
"train_runtime": 7716.3929,
"train_samples": 7500,
"train_samples_per_second": 0.972,
"train_steps_per_second": 0.061
}