Llama-3.1-8B-medquad-V1 / all_results.json
mariamoracrossitcr's picture
End of training
d06e2db verified
raw
history blame contribute delete
364 Bytes
{
"epoch": 1.988950276243094,
"eval_loss": 0.9016749262809753,
"eval_runtime": 263.4205,
"eval_samples_per_second": 5.493,
"eval_steps_per_second": 0.687,
"total_flos": 2.1394363129941197e+17,
"train_loss": 0.9953938881556194,
"train_runtime": 10885.5423,
"train_samples_per_second": 7.444,
"train_steps_per_second": 0.039
}