reasoning_6_chat / train_results.json
sedrickkeh's picture
End of training
c38f322 verified
{
"epoch": 4.994311717861206,
"total_flos": 3.168901699486089e+18,
"train_loss": 0.34217921912262816,
"train_runtime": 81219.8428,
"train_samples_per_second": 3.463,
"train_steps_per_second": 0.027
}