Qwen-2.5-7B-Simple-RL / train_results.json
namazifar's picture
Model save
1014960 verified
{
"total_flos": 0.0,
"train_loss": 0.05658998479865941,
"train_runtime": 6508.7782,
"train_samples": 7500,
"train_samples_per_second": 1.152,
"train_steps_per_second": 0.009
}