OpenR1-Qwen-7B-Math-repeat-B2048 / train_results.json
Rano23's picture
Model save
cb6a0f6 verified
{
"total_flos": 2.75595486135373e+18,
"train_loss": 0.411641573528208,
"train_runtime": 64602.1937,
"train_samples": 93733,
"train_samples_per_second": 0.797,
"train_steps_per_second": 0.05
}