Update README.md
Browse files
README.md
CHANGED
|
@@ -4,21 +4,54 @@ language:
|
|
| 4 |
- en
|
| 5 |
library_name: transformers
|
| 6 |
pipeline_tag: text-generation
|
| 7 |
-
---
|
| 8 |
-
9o
|
| 9 |
-
---
|
| 10 |
-
license:
|
| 11 |
-
base_model:
|
| 12 |
tags:
|
| 13 |
- merge
|
| 14 |
- mergekit
|
| 15 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
---
|
| 17 |
|
|
|
|
|
|
|
| 18 |
# Maxine-34B-stock
|
| 19 |
|
| 20 |
Maxine-34B-stock is a merge of the following models:
|
| 21 |
|
|
|
|
|
|
|
| 22 |
## Configuration
|
| 23 |
|
| 24 |
```yaml
|
|
@@ -53,4 +86,21 @@ pipeline = transformers.pipeline(
|
|
| 53 |
|
| 54 |
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 55 |
print(outputs[0]["generated_text"])
|
| 56 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
- en
|
| 5 |
library_name: transformers
|
| 6 |
pipeline_tag: text-generation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
tags:
|
| 8 |
- merge
|
| 9 |
- mergekit
|
| 10 |
+
- ConvexAI/Luminex-34B-v0.2
|
| 11 |
+
- fblgit/UNA-34BeagleSimpleMath-32K-v1
|
| 12 |
+
- chemistry
|
| 13 |
+
- biology
|
| 14 |
+
- math
|
| 15 |
+
base_model:
|
| 16 |
+
- ConvexAI/Luminex-34B-v0.2
|
| 17 |
+
- fblgit/UNA-34BeagleSimpleMath-32K-v1
|
| 18 |
+
model-index:
|
| 19 |
+
- name: Maxine-34B-stock
|
| 20 |
+
results:
|
| 21 |
+
- task:
|
| 22 |
+
type: text-generation
|
| 23 |
+
metrics:
|
| 24 |
+
- name: Average
|
| 25 |
+
type: Average
|
| 26 |
+
value: 77.28
|
| 27 |
+
- name: ARC
|
| 28 |
+
type: ARC
|
| 29 |
+
value: 74.06
|
| 30 |
+
- name: GSM8K
|
| 31 |
+
type: GSM8K
|
| 32 |
+
value: 72.18
|
| 33 |
+
- name: Winogrande
|
| 34 |
+
type: Winogrande
|
| 35 |
+
value: 83.9
|
| 36 |
+
- name: TruthfulQA
|
| 37 |
+
type: TruthfulQA
|
| 38 |
+
value: 70.18
|
| 39 |
+
- name: HellaSwag
|
| 40 |
+
type: HellaSwag
|
| 41 |
+
value: 86.74
|
| 42 |
+
source:
|
| 43 |
+
name: Open LLM Leaderboard
|
| 44 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
|
| 45 |
---
|
| 46 |
|
| 47 |
+
<center><img src='https://i.imgur.com/dU9dUh0.png' width='500px'></center>
|
| 48 |
+
|
| 49 |
# Maxine-34B-stock
|
| 50 |
|
| 51 |
Maxine-34B-stock is a merge of the following models:
|
| 52 |
|
| 53 |
+
**04-07-2024 - To date, louisbrulenaudet/Maxine-34B-stock is the "Best 🤝 base merges and moerges model of around 30B" on the Open LLM Leaderboard.**
|
| 54 |
+
|
| 55 |
## Configuration
|
| 56 |
|
| 57 |
```yaml
|
|
|
|
| 86 |
|
| 87 |
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 88 |
print(outputs[0]["generated_text"])
|
| 89 |
+
```
|
| 90 |
+
|
| 91 |
+
## Citing & Authors
|
| 92 |
+
|
| 93 |
+
If you use this code in your research, please use the following BibTeX entry.
|
| 94 |
+
|
| 95 |
+
```BibTeX
|
| 96 |
+
@misc{louisbrulenaudet2024,
|
| 97 |
+
author = {Louis Brulé Naudet},
|
| 98 |
+
title = {Maxine-34B-stock, an xtraordinary 34B model},
|
| 99 |
+
year = {2024}
|
| 100 |
+
howpublished = {\url{https://huggingface.co/louisbrulenaudet/Maxine-34B-stock}},
|
| 101 |
+
}
|
| 102 |
+
```
|
| 103 |
+
|
| 104 |
+
## Feedback
|
| 105 |
+
|
| 106 |
+
If you have any feedback, please reach out at [[email protected]](mailto:[email protected]).
|