quip-qwen2 / train_results.json
Anis1123's picture
Initial model upload
6782b24 verified
raw
history blame contribute delete
No virus
260 Bytes
{
"epoch": 2.9936575052854124,
"num_input_tokens_seen": 4597544,
"total_flos": 5.475245368337203e+16,
"train_loss": 2.9544576898133013,
"train_runtime": 3114.1126,
"train_samples_per_second": 1.822,
"train_steps_per_second": 0.114
}