hp_ablations_qwen_adambeta1_0.95 / train_results.json
sedrickkeh's picture
End of training
9f3bb40 verified
raw
history blame
219 Bytes
{
"epoch": 2.996011396011396,
"total_flos": 2755219238682624.0,
"train_loss": 0.6039808779909549,
"train_runtime": 71394.4326,
"train_samples_per_second": 9.439,
"train_steps_per_second": 0.018
}