DeepSeek-R1-Distill-Qwen-7B-GRPO / train_results.json
PrismXu's picture
Model save
22ab92a verified
raw
history blame contribute delete
182 Bytes
{
"total_flos": 0.0,
"train_loss": 0.0,
"train_runtime": 9046.809,
"train_samples": 1250,
"train_samples_per_second": 0.276,
"train_steps_per_second": 0.002
}