Files
DeepSeek-R1-Distill-Qwen-1.…/all_results.json

8 lines
213 B
JSON
Raw Normal View History

{
"total_flos": 287156144701440.0,
"train_loss": 0.4225423020029825,
"train_runtime": 9655.7571,
"train_samples": 18521,
"train_samples_per_second": 2.082,
"train_steps_per_second": 0.033
}