ViperEkura a7bbc7b29f fix: align SFT/DPO figures and text with actual training data ...

- SFT: 1,000 steps/WSD → ~3,800 steps/cosine; loss ~2.5→1.6 → ~2.1→1.5
- DPO: fix preference-loss narrative to match training-loss curve;
  correct initial grad-norm (~50 → ~200)
- Table 2: peak LR 1.5e-4 → 2.0e-4 (matches pt_metric.png)
- Clarify scheduling: pretraining=WSD, SFT+DPO=cosine
- Add disclaimer to ckpt_weight_density_per_run caption for legacy iter labels

2026-07-24 06:15:42 +08:00

84 KiB

1500x380px

Raw History