Update README.md
Browse files
README.md
CHANGED
@@ -110,13 +110,13 @@ More evaluations are under way.
|
|
110 |
|
111 |
### Accuracy
|
112 |
|
113 |
-
| | Recovery (%)
|
114 |
-
| ---------------------------------------------- | :-----------: |
|
115 |
-
| ARC-Challenge<br>25-shot |
|
116 |
-
| GSM8k<br>5-shot |
|
117 |
-
| HellaSwag<br>10-shot |
|
118 |
-
| MMLU<br>5-shot |
|
119 |
-
| TruthfulQA<br>0-shot |
|
120 |
-
| WinoGrande<br>5-shot |
|
121 |
-
| **OpenLLM v1<br>Average Score**
|
122 |
|
|
|
110 |
|
111 |
### Accuracy
|
112 |
|
113 |
+
| | Recovery (%) | meta-llama/Llama-4-Maverick-17B-128E-Instruct | RedHatAI/Llama-4-Maverick-17B-128E-Instruct-quantized.w4a16<br>(this model) |
|
114 |
+
| ---------------------------------------------- | :-----------: | :-------------------------------------------: | :-----------------------------------------------------------------: |
|
115 |
+
| ARC-Challenge<br>25-shot | 96.6 | 73.55 | 71.08 |
|
116 |
+
| GSM8k<br>5-shot | 99.7 | 93.18 | 92.87 |
|
117 |
+
| HellaSwag<br>10-shot | 99.6 | 87.27 | 86.95 |
|
118 |
+
| MMLU<br>5-shot | 99.8 | 85.98 | 85.78 |
|
119 |
+
| TruthfulQA<br>0-shot | 100.0 | 62.81 | 62.85 |
|
120 |
+
| WinoGrande<br>5-shot | 100.5 | 78.53 | 78.93 |
|
121 |
+
| **OpenLLM v1<br>Average Score** | **99.4** | **80.22** | **79.74** |
|
122 |
|