llama-3_1-8b-overfit-ua / all_results.json
antonpolishko's picture
Model save
974e73b verified
raw
history blame contribute delete
235 Bytes
{
"epoch": 3.0,
"total_flos": 6.427401199279931e+18,
"train_loss": 1.711617823146263,
"train_runtime": 5145.0339,
"train_samples": 95663,
"train_samples_per_second": 13.545,
"train_steps_per_second": 0.212
}