Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,11 @@ This is a experimental MoE created using Mergekit from
|
|
| 18 |
* [NousResearch/Hermes-2-Theta-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B)
|
| 19 |
* [rombodawg/Llama-3-8B-Instruct-Coder](https://huggingface.co/rombodawg/Llama-3-8B-Instruct-Coder)
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
Mergekit yaml file:
|
| 22 |
```
|
| 23 |
base_model: Meta-Llama-3-8B-Instruct
|
|
|
|
| 18 |
* [NousResearch/Hermes-2-Theta-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B)
|
| 19 |
* [rombodawg/Llama-3-8B-Instruct-Coder](https://huggingface.co/rombodawg/Llama-3-8B-Instruct-Coder)
|
| 20 |
|
| 21 |
+
Evaluation:
|
| 22 |
+
Q4_K_M:
|
| 23 |
+
* GSM8K (5-shot): 0.6983 ± 0.0126
|
| 24 |
+
* GSM8K (8-shot, cot): 0.674 ± 0.0129
|
| 25 |
+
|
| 26 |
Mergekit yaml file:
|
| 27 |
```
|
| 28 |
base_model: Meta-Llama-3-8B-Instruct
|