Update README.md
Browse files
README.md
CHANGED
|
@@ -38,7 +38,28 @@ The `Locutusque/Hyperion-3.0-Mistral-7B-alpha` model was fine-tuned on 200,000 e
|
|
| 38 |
Coming Soon
|
| 39 |
|
| 40 |
## Evaluation Results
|
| 41 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
|
| 43 |
## How to Use
|
| 44 |
```python
|
|
|
|
| 38 |
Coming Soon
|
| 39 |
|
| 40 |
## Evaluation Results
|
| 41 |
+
Zero-shot AGIEval
|
| 42 |
+
|
| 43 |
+
| Tasks |Version|Filter|n-shot| Metric |Value | |Stderr|
|
| 44 |
+
|---------------------------------|-------|------|------|--------|-----:|---|-----:|
|
| 45 |
+
|agieval_nous |N/A |none |None |acc_norm|0.3500|± |0.0092|
|
| 46 |
+
| | |none |None |acc |0.3519|± |0.0093|
|
| 47 |
+
| - agieval_aqua_rat | 1|none |None |acc |0.2087|± |0.0255|
|
| 48 |
+
| | |none |None |acc_norm|0.2165|± |0.0259|
|
| 49 |
+
| - agieval_logiqa_en | 1|none |None |acc |0.3164|± |0.0182|
|
| 50 |
+
| | |none |None |acc_norm|0.3533|± |0.0187|
|
| 51 |
+
| - agieval_lsat_ar | 1|none |None |acc |0.2348|± |0.0280|
|
| 52 |
+
| | |none |None |acc_norm|0.2174|± |0.0273|
|
| 53 |
+
| - agieval_lsat_lr | 1|none |None |acc |0.3529|± |0.0212|
|
| 54 |
+
| | |none |None |acc_norm|0.3647|± |0.0213|
|
| 55 |
+
| - agieval_lsat_rc | 1|none |None |acc |0.4721|± |0.0305|
|
| 56 |
+
| | |none |None |acc_norm|0.4201|± |0.0301|
|
| 57 |
+
| - agieval_sat_en | 1|none |None |acc |0.6019|± |0.0342|
|
| 58 |
+
| | |none |None |acc_norm|0.6117|± |0.0340|
|
| 59 |
+
| - agieval_sat_en_without_passage| 1|none |None |acc |0.4078|± |0.0343|
|
| 60 |
+
| | |none |None |acc_norm|0.3835|± |0.0340|
|
| 61 |
+
| - agieval_sat_math | 1|none |None |acc |0.3091|± |0.0312|
|
| 62 |
+
| | |none |None |acc_norm|0.2364|± |0.0287|
|
| 63 |
|
| 64 |
## How to Use
|
| 65 |
```python
|