openr1_codeforces_1k / train_results.json
ryanmarten's picture
End of training
5b41988 verified
{
"epoch": 6.72,
"total_flos": 2.46611802945749e+17,
"train_loss": 0.6276680737733841,
"train_runtime": 7354.819,
"train_samples_per_second": 0.952,
"train_steps_per_second": 0.01
}