Ubuntu
Added trained v1.6 model
d151e03
{
"epoch": 0.96,
"num_input_tokens_seen": 45448,
"total_flos": 1891508470824960.0,
"train_loss": 1.1797900597254436,
"train_runtime": 413.707,
"train_samples_per_second": 0.242,
"train_steps_per_second": 0.029
}