Mixtral-8x7B-v0.1_case-briefs / train_results.json
zlucia's picture
End of training
f9e757f verified
raw
history blame
166 Bytes
{
"epoch": 2.0,
"train_loss": 0.9311216080269846,
"train_runtime": 4522.2924,
"train_samples_per_second": 1.042,
"train_steps_per_second": 0.065
}