Model_1B_Bush / train_results.json
GPT-JF's picture
End of training
833232a
raw
history blame contribute delete
192 Bytes
{
"epoch": 5.0,
"train_loss": 2.4309686013099934,
"train_runtime": 1406.649,
"train_samples": 2566,
"train_samples_per_second": 9.121,
"train_steps_per_second": 2.282
}