Files
SmolLM2-1.7B-16k-SFT-Tulu3-…/all_results.json

8 lines
217 B
JSON
Raw Normal View History

{
"total_flos": 3199489317601280.0,
"train_loss": 0.7138549819888028,
"train_runtime": 48209.3705,
"train_samples": 936509,
"train_samples_per_second": 38.852,
"train_steps_per_second": 0.304
}