MiniLMv2-L6-H384_R-fineweb-100k / train_results.json
pszemraj's picture
End of training
4cefba7
raw
history blame contribute delete
No virus
331 Bytes
{
"epoch": 1.9998993609419817,
"num_input_tokens_seen": 162790400,
"total_flos": 1.059416318592e+16,
"train_loss": 4.377665276304727,
"train_runtime": 2972.3443,
"train_samples": 158982,
"train_samples_per_second": 106.974,
"train_steps_per_second": 0.836,
"train_tokens_per_second": 54770.764
}