gpt2-xl-lora-multi / train_results.json
MHGanainy's picture
MHGanainy/gpt2-xl-lora-multi
ac813e8 verified
{
"epoch": 1.0,
"total_flos": 7.322050088623145e+18,
"train_loss": 2.3946414439395287,
"train_runtime": 14836.2948,
"train_samples_per_second": 52.513,
"train_steps_per_second": 3.282
}