Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
Kurt232's picture
Model save
1b83f39 verified
raw
history blame
200 Bytes
{
"total_flos": 0.0,
"train_loss": 2.1605342640346974,
"train_runtime": 168649.3171,
"train_samples": 72441,
"train_samples_per_second": 0.43,
"train_steps_per_second": 0.003
}