Files
nemotron-1000-opt1k__Qwen3-8B/train_results.json

12 lines
396 B
JSON
Raw Normal View History

{
"achieved_tflops_per_gpu": 5273.804457747945,
"achieved_tflops_per_gpu_theoretical": 152245.65524414752,
"epoch": 7.0,
"mfu_percent": 372.70702881610924,
"mfu_percent_theoretical": 10759.410264604065,
"total_flos": 2.891732460272353e+17,
"train_loss": 0.0,
"train_runtime": 1.7135,
"train_samples_per_second": 4085.295,
"train_steps_per_second": 44.938
}