Qwen-2.5-3B-Simple-RL / all_results.json
Typiiing's picture
Model save
dcc8c2d verified
raw
history blame contribute delete
201 Bytes
{
"total_flos": 0.0,
"train_loss": 0.002609439611100802,
"train_runtime": 17087.0679,
"train_samples": 7500,
"train_samples_per_second": 0.439,
"train_steps_per_second": 0.018
}