Qwen-2.5-7B-Simple-RL / train_results.json
Blancy's picture
Model save
bbe1200 verified
{
"total_flos": 0.0,
"train_loss": 0.0006319970209619013,
"train_runtime": 19204.1352,
"train_samples": 10000,
"train_samples_per_second": 0.521,
"train_steps_per_second": 0.004
}