Update README.md
Browse files
README.md
CHANGED
|
@@ -21,6 +21,10 @@ tags:
|
|
| 21 |
|
| 22 |
phixtral-4x2_8 is the first Mixure of Experts (MoE) made with four [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) models, inspired by the [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1) architecture. It performs better than each individual expert.
|
| 23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
## 🏆 Evaluation
|
| 25 |
|
| 26 |
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|
|
|
|
| 21 |
|
| 22 |
phixtral-4x2_8 is the first Mixure of Experts (MoE) made with four [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) models, inspired by the [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1) architecture. It performs better than each individual expert.
|
| 23 |
|
| 24 |
+
## ⚡ Quantized models
|
| 25 |
+
|
| 26 |
+
* **GPTQ**: https://huggingface.co/TheBloke/phixtral-4x2_8-GPTQ
|
| 27 |
+
|
| 28 |
## 🏆 Evaluation
|
| 29 |
|
| 30 |
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|