deepseek-llm-7b-chat-sa-v0.1 / train_results.json
sci-m-wang's picture
Upload 13 files
740432a verified
{
"epoch": 4.9976558837318334,
"total_flos": 1.9473516408636703e+18,
"train_loss": 0.5572177461119575,
"train_runtime": 35175.961,
"train_samples_per_second": 1.213,
"train_steps_per_second": 0.076
}