Files
Llama-3.1-8B-Instruct_SFT_m…/train_results.json

9 lines
257 B
JSON
Raw Permalink Normal View History

{
"ewc_loss": 2.968311309814453e-05,
"total_flos": 5.628159003328302e+19,
"train_loss": 0.4093159021503956,
"train_runtime": 46764.448,
"train_samples": 125770,
"train_samples_per_second": 8.068,
"train_steps_per_second": 0.504
}