1b_distill_width_prune / train_results.json
friendshipkim's picture
Model save
9a0f1ca verified
raw
history blame
243 Bytes
{
"epoch": 0.08267797093601252,
"total_flos": 0.0,
"train_loss": 6.551032936291113e-05,
"train_runtime": 86.8457,
"train_samples": 15482525,
"train_samples_per_second": 14738.777,
"train_steps_per_second": 230.293
}