llama3-8b-summarize-gpt4o-128k / train_results.json
chansung's picture
Model save
b0bd756 verified
raw
history blame contribute delete
251 Bytes
{
"epoch": 9.954337899543379,
"total_flos": 6.456679991336763e+18,
"train_loss": 0.7130365929472338,
"train_runtime": 9668.9725,
"train_samples": 129221,
"train_samples_per_second": 14.452,
"train_steps_per_second": 0.113
}