DeepSeek-R1-Distill-Qwen-1.5B-S / train_results.json
leonMW's picture
Model save
4783a3f verified
{
"total_flos": 0.0,
"train_loss": 0.002908330309488203,
"train_runtime": 11900.5972,
"train_samples": 3000,
"train_samples_per_second": 0.504,
"train_steps_per_second": 0.015
}