Qwen-2.5-7B-Simple-RL / train_results.json
dadadar's picture
Model save
c0c3d48 verified
{
"total_flos": 0.0,
"train_loss": 0.03916507870213598,
"train_runtime": 36868.5034,
"train_samples": 7500,
"train_samples_per_second": 0.203,
"train_steps_per_second": 0.003
}