HuanxinSheng's picture
robojudge-4b-v02 SFT final (step 5780)
fac5b68 verified
|
Raw History Blame Contribute Delete
534 Bytes
base rows 231, new rows 231, paired 231
| slice | n | base acc | new acc | Δacc [95% CI] | base NLL | new NLL | ΔNLL [95% CI] | new-only/base-only correct |
|---|---|---|---|---|---|---|---|---|
| **variant=clean** | 231 | 77.92 | 77.92 | +0.00 [-3.10, +2.64] | 0.482 | 0.547 | +0.065 [-0.002, +0.154] | 5/5 |
| task=extraction | 24 | 100.00 | 100.00 | +0.00 [+0.00, +0.00] | 0.004 | 0.002 | -0.002 [-0.005, -0.000] | 0/0 |
| task=intent | 24 | 95.83 | 95.83 | +0.00 [+0.00, +0.00] | 0.061 | 0.060 | -0.001 [-0.003, -0.000] | 0/0 |