Update README.md
Browse files
README.md
CHANGED
|
@@ -31,8 +31,6 @@ model-index:
|
|
| 31 |
|
| 32 |
<p><h1> speechless-mistral-six-in-one-7b </h1></p>
|
| 33 |
|
| 34 |
-
# JUST for TEST!
|
| 35 |
-
|
| 36 |
This model is a merge of 6 SOTA Mistral-7B based models:
|
| 37 |
- ehartford/dolphin-2.1-mistral-7b
|
| 38 |
- Open-Orca/Mistral-7B-OpenOrca
|
|
@@ -42,6 +40,20 @@ This model is a merge of 6 SOTA Mistral-7B based models:
|
|
| 42 |
- HuggingFaceH4/zephyr-7b-alpha
|
| 43 |
|
| 44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
## HumanEval
|
| 46 |
|
| 47 |
| Metric | Value |
|
|
|
|
| 31 |
|
| 32 |
<p><h1> speechless-mistral-six-in-one-7b </h1></p>
|
| 33 |
|
|
|
|
|
|
|
| 34 |
This model is a merge of 6 SOTA Mistral-7B based models:
|
| 35 |
- ehartford/dolphin-2.1-mistral-7b
|
| 36 |
- Open-Orca/Mistral-7B-OpenOrca
|
|
|
|
| 40 |
- HuggingFaceH4/zephyr-7b-alpha
|
| 41 |
|
| 42 |
|
| 43 |
+
[Model benchmark](https://huggingface.co/uukuguy/speechless-mistral-six-in-one-7b/discussions/1) by [sethuiyer](https://huggingface.co/sethuiyer) . Thanks a lot.
|
| 44 |
+
> I tested the Q6_0 version of the model against LLaMa2 70B chat and here are the results - Scoring as per ChatGPT and Bard's average. Named this model Mixtral. Questions taken from MT-Benchmark.
|
| 45 |
+
>
|
| 46 |
+
> On a scale of 0 to 100, I would rate Mixtral at 98. Here's why:
|
| 47 |
+
>
|
| 48 |
+
> - Intellect (100/100) - Mixtral has demonstrated immense intellectual abilities through its comprehensive knowledge and logical reasoning skills.
|
| 49 |
+
> - Creativity (98/100) - In addition to being highly intelligent, Mixtral also displays impressive creative talents through its unique, nuanced responses.
|
| 50 |
+
> - Adaptability (98/100) - Mixtral can converse flexibly on a wide variety of topics, adapting smoothly based on contextual cues.
|
| 51 |
+
> - Communication (97/100) - Mixtral communicates clearly and eloquently through written language, thoroughly answering questions.
|
| 52 |
+
> - Problem-Solving (98/100) - Questions are addressed comprehensively, considering multiple perspectives to arrive at well-thought solutions.
|
| 53 |
+
> - Personability (97/100) - Responses are warm, inviting and non-threatening due to Mixtral's kindness and thoughtfulness.
|
| 54 |
+
>
|
| 55 |
+
> Overall, a very capable model for it's size.
|
| 56 |
+
|
| 57 |
## HumanEval
|
| 58 |
|
| 59 |
| Metric | Value |
|