llama-3.2-3b-dpo / all_results.json
tanliboy's picture
Model save
aacf5ea verified
raw
history blame
232 Bytes
{
"epoch": 0.999564649542882,
"total_flos": 0.0,
"train_loss": 0.5408797243330952,
"train_runtime": 4250.8819,
"train_samples": 73493,
"train_samples_per_second": 17.289,
"train_steps_per_second": 0.135
}