Files
Llama-3.1-8B-Instruct_SFT_s…/all_results.json

9 lines
260 B
JSON
Raw Permalink Normal View History

{
"ewc_loss": 1.055002212524414e-05,
"total_flos": 3.1829322264618205e+19,
"train_loss": 1.0248253059881185,
"train_runtime": 37191.1487,
"train_samples": 145693,
"train_samples_per_second": 11.752,
"train_steps_per_second": 0.735
}