zephyr-7b-dpo-qlora-8e0975a / train_results.json
lewtun's picture
lewtun HF staff
Duplicate from alignment-handbook/zephyr-7b-dpo-qlora
f1395af
raw
history blame contribute delete
193 Bytes
{
"epoch": 1.0,
"train_loss": 0.583915277301329,
"train_runtime": 6210.8046,
"train_samples": 61135,
"train_samples_per_second": 9.843,
"train_steps_per_second": 0.154
}