Qwen-2.5-7B-GRPO-Base-4Action_774 / train_results.json
luckeciano's picture
Model save
69cb3d1 verified
raw
history blame contribute delete
202 Bytes
{
"total_flos": 0.0,
"train_loss": 8.040418569832397e-10,
"train_runtime": 17425.8032,
"train_samples": 7500,
"train_samples_per_second": 0.551,
"train_steps_per_second": 0.006
}