zephyr-7b-dpo-lora-r16-20k / train_results.json
LaoRay's picture
Model save
32da1fc verified
raw
history blame
218 Bytes
{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.5874967678070069,
"train_runtime": 15864.6959,
"train_samples": 20000,
"train_samples_per_second": 1.261,
"train_steps_per_second": 0.079
}