Llama-2-7b-Llama-3.1-8B-mix / train_results.json
AmberYifan's picture
Model save
bd209ea verified
raw
history blame
216 Bytes
{
"epoch": 3.0,
"total_flos": 0.0,
"train_loss": 0.6005911044973545,
"train_runtime": 1656.4163,
"train_samples": 2000,
"train_samples_per_second": 3.622,
"train_steps_per_second": 0.114
}