qwen-2.5-3b-r1-countdown / train_results.json
sciarrilli's picture
Model save
112c526 verified
{
"total_flos": 0.0,
"train_loss": 0.0009040301745537565,
"train_runtime": 63789.5966,
"train_samples": 45000,
"train_samples_per_second": 3.16,
"train_steps_per_second": 0.007
}