Files
Qwen3-8B-fim-v2v3pt-swe-leg…/train_results.json

8 lines
210 B
JSON
Raw Normal View History

{
"epoch": 4.0,
"total_flos": 7.881538624284525e+18,
"train_loss": 0.13219511501680492,
"train_runtime": 266794.0573,
"train_samples_per_second": 0.212,
"train_steps_per_second": 0.004
}