hp_ablations_gemma_adambeta2_0.99 / train_results.json
sedrickkeh's picture
End of training
5b2df1b verified
raw
history blame contribute delete
220 Bytes
{
"epoch": 2.9991537376586743,
"total_flos": 5064195066298368.0,
"train_loss": 0.5476742999907093,
"train_runtime": 122403.0515,
"train_samples_per_second": 5.56,
"train_steps_per_second": 0.011
}