DeepSeek-R1-Distill-Qwen-14B-S / train_results.json
leonMW's picture
Model save
fb78aae verified
{
"total_flos": 0.0,
"train_loss": 0.004917917826080359,
"train_runtime": 17643.0116,
"train_samples": 2027,
"train_samples_per_second": 0.23,
"train_steps_per_second": 0.029
}