Update README.md
Browse files
README.md
CHANGED
|
@@ -42,17 +42,15 @@ remaining layers with a new classification head.
|
|
| 42 |
# NLI Evaluation Results
|
| 43 |
|
| 44 |
F1-Micro scores (equivalent to accuracy) for each dataset.
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
|
| 48 |
-
|
|
| 49 |
-
| `
|
| 50 |
-
| `dleemiller/
|
| 51 |
-
| `tasksource/ModernBERT-
|
| 52 |
-
| `dleemiller/ModernCE-
|
| 53 |
-
| `
|
| 54 |
-
| `dleemiller/EttinX-nli-xs` | 0.7013 | 0.8376 | 0.8380 | 0.8979 | 0.2780 | 0.2840 | 0.2800 | 0.5838 | 0.7521 |
|
| 55 |
-
| `dleemiller/EttinX-nli-xxs` | 0.6842 | 0.7988 | 0.8047 | 0.8851 | 0.2590 | 0.3060 | 0.2992 | 0.5426 | 0.7018 |
|
| 56 |
|
| 57 |
|
| 58 |
---
|
|
|
|
| 42 |
# NLI Evaluation Results
|
| 43 |
|
| 44 |
F1-Micro scores (equivalent to accuracy) for each dataset.
|
| 45 |
+
Performance was measured at bs=32 using a Nvidia Blackwell PRO 6000 Max-Q.
|
| 46 |
+
|
| 47 |
+
| Model | finecat | mnli | mnli_mismatched | snli | anli_r1 | anli_r2 | anli_r3 | wanli | lingnli | Throughput (samples/s) | Peak GPU Mem (MB) |
|
| 48 |
+
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
| 49 |
+
| `MoritzLaurer/DeBERTa-v3-large-mnli-fever-anli-ling-wanli` | **0.8233** | <u>0.9121</u> | 0.9079 | 0.8898 | **0.7960** | **0.6830** | **0.6400** | <u>0.7700</u> | **0.8821** | 454.96 | 3250.44 |
|
| 50 |
+
| `dleemiller/finecat-nli-l` | <u>0.8227</u> | **0.9152** | **0.9265** | 0.9162 | <u>0.7480</u> | <u>0.5700</u> | <u>0.5433</u> | **0.7706** | <u>0.8742</u> | 539.04 | 1838.06 |
|
| 51 |
+
| `tasksource/ModernBERT-large-nli` | 0.7959 | 0.8983 | <u>0.9229</u> | 0.9188 | 0.7260 | 0.5110 | 0.4925 | 0.6978 | 0.8504 | 543.44 | 1838.06 |
|
| 52 |
+
| `dleemiller/ModernCE-large-nli` | 0.7811 | 0.9088 | 0.9205 | **0.9273** | 0.6630 | 0.4860 | 0.4408 | 0.6576 | 0.8566 | 540.74 | 1838.06 |
|
| 53 |
+
| `cross-encoder/nli-deberta-v3-large` | 0.7618 | 0.9019 | 0.9049 | <u>0.9220</u> | 0.5300 | 0.4170 | 0.3758 | 0.6548 | 0.8466 | 448.35 | 3250.44 |
|
|
|
|
|
|
|
| 54 |
|
| 55 |
|
| 56 |
---
|