OpenO1-LLama-8B-v0.1 / all_results.json
paralym's picture
add sft model
49758f8
raw
history blame contribute delete
345 Bytes
{
"epoch": 3.0,
"eval_loss": 0.6375684142112732,
"eval_runtime": 168.7291,
"eval_samples_per_second": 30.807,
"eval_steps_per_second": 3.852,
"total_flos": 391576352194560.0,
"train_loss": 0.44177393167463547,
"train_runtime": 22000.6442,
"train_samples_per_second": 6.378,
"train_steps_per_second": 0.1
}