Update README.md
Browse files
README.md
CHANGED
@@ -100,12 +100,12 @@ The dataset is comprised of a mixture of open datasets large-scale datasets avai
|
|
100 |
## Other benchmark:
|
101 |
| Metric | Value |
|
102 |
|-----------------------|---------------------------|
|
103 |
-
| ARC (25-shot) | 0
|
104 |
-
| HellaSwag (10-shot) |
|
105 |
-
| MMLU (5-shot) |
|
106 |
| TruthfulQA (0-shot) | |
|
107 |
-
| Winogrande (5-shot) |
|
108 |
-
| GSM8K (5-shot) |
|
109 |
|
110 |
|
111 |
- BigBench: 0.3526.
|
|
|
100 |
## Other benchmark:
|
101 |
| Metric | Value |
|
102 |
|-----------------------|---------------------------|
|
103 |
+
| ARC (25-shot) | 47.0 |
|
104 |
+
| HellaSwag (10-shot) | 74.2 |
|
105 |
+
| MMLU (5-shot) | 46.3 |
|
106 |
| TruthfulQA (0-shot) | |
|
107 |
+
| Winogrande (5-shot) | 65.5 |
|
108 |
+
| GSM8K (5-shot) | 42.3 |
|
109 |
|
110 |
|
111 |
- BigBench: 0.3526.
|