gpt2-xl-lora-multi-5 / train_results.json
MHGanainy's picture
MHGanainy/gpt2-xl-lora-multi-5
3515a7e verified
{
"epoch": 1.0,
"total_flos": 8.917940750474281e+17,
"train_loss": 2.4952426230985356,
"train_runtime": 1729.9843,
"train_samples_per_second": 56.6,
"train_steps_per_second": 3.538
}