llama3-1b-summarize-gpt4o-128k / train_results.json
chansung's picture
Model save
e4e88e3 verified
raw
history blame contribute delete
252 Bytes
{
"epoch": 9.654545454545454,
"total_flos": 8.156088875152835e+17,
"train_loss": 1.7710220513520418,
"train_runtime": 1245.0854,
"train_samples": 129221,
"train_samples_per_second": 112.233,
"train_steps_per_second": 0.217
}