Llama-3.1-8B-Llama-2-7b-mix / train_results.json
AmberYifan's picture
Model save
bb34263 verified
raw
history blame
217 Bytes
{
"epoch": 3.0,
"total_flos": 0.0,
"train_loss": 0.4980044298913405,
"train_runtime": 4849.4416,
"train_samples": 10000,
"train_samples_per_second": 6.186,
"train_steps_per_second": 0.194
}