tFINE-850m-24x24-v0.5-instruct-L1 / train_results.json
pszemraj's picture
End of training
3b6f2da verified
raw
history blame contribute delete
295 Bytes
{
"epoch": 0.9999924371336737,
"num_input_tokens_seen": 435513684,
"total_flos": 2.1022605922963784e+18,
"train_loss": 1.4536694422503678,
"train_runtime": 158351.1039,
"train_samples": 1586697,
"train_samples_per_second": 10.02,
"train_steps_per_second": 0.078
}