hp_ablations_qwen_adambeta1_0.92 / train_results.json
sedrickkeh's picture
End of training
89e23e2 verified
raw
history blame contribute delete
218 Bytes
{
"epoch": 2.996011396011396,
"total_flos": 2755219238682624.0,
"train_loss": 0.6040076841139539,
"train_runtime": 71054.678,
"train_samples_per_second": 9.484,
"train_steps_per_second": 0.018
}