Qwen2.5-1.5B-Open-R1-Math-GRPO / train_results.json
jonatatyska's picture
Model save
be29d92 verified
raw
history blame contribute delete
200 Bytes
{
"total_flos": 0.0,
"train_loss": -0.11113634665512986,
"train_runtime": 2287.8706,
"train_samples": 1600,
"train_samples_per_second": 0.699,
"train_steps_per_second": 0.025
}