Qwen-2.5-7B-GRPO-NoBaseline_951 / train_results.json
luckeciano's picture
Model save
0012354 verified
raw
history blame contribute delete
200 Bytes
{
"total_flos": 0.0,
"train_loss": -0.7403125202655793,
"train_runtime": 17408.0197,
"train_samples": 7500,
"train_samples_per_second": 0.551,
"train_steps_per_second": 0.006
}