alvarobartt HF staff commited on
Commit
4b5d85b
·
1 Parent(s): 4029d4c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +60 -0
README.md CHANGED
@@ -1,3 +1,63 @@
1
  ---
 
 
 
 
 
 
 
2
  license: mit
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ pipeline_tag: text-generation
3
+ inference: false
4
+ tags:
5
+ - notus
6
+ - mlx
7
+ language:
8
+ - en
9
  license: mit
10
+ library_name: mlx
11
  ---
12
+
13
+ # Notus 7B v1 (✨ 4-bit)
14
+
15
+ <div align="center">
16
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/60f0608166e5701b80ed3f02/dj-spsk9eXMMXVGxK6jRz.png" alt="A banner representing Notus, the wind god of the south, in a mythical and artistic style. The banner features a strong, swirling breeze, embodying the warm, wet character of the southern wind. Gracefully flowing across the scene are several paper planes, caught in the gentle yet powerful gusts of Notus. The background is a blend of warm colors, symbolizing the heat of the south, with hints of blue and green to represent the moisture carried by this wind. The overall atmosphere is one of dynamic movement and warmth."/>
17
+ </div>
18
+
19
+ Notus is a collection of fine-tuned models using Direct Preference Optimization (DPO) and related RLHF techniques following a data-first approach. This model is the first version, fine-tuned with DPO on top of [`alignment-handbook/zephyr-7b-sft-full`](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full), which is the SFT model produced to create [`HuggingFaceH4/zephyr-7b-beta`](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta).
20
+
21
+ More information at [argilla/notus-7b-v1](https://huggingface.co/argilla/notus-7b-v1)
22
+
23
+ This repository contains the `notus-7b-v1` weights in `npz` format in 4-bit suitable for use with Apple's MLX framework (from 0.6.0 onwards).
24
+
25
+ ## Use with MLX
26
+
27
+ ```bash
28
+ pip install mlx
29
+ pip install huggingface_hub hf_transfer
30
+ git clone https://github.com/ml-explore/mlx-examples.git
31
+ cd mlx-examples
32
+
33
+ # Download model
34
+ export HF_HUB_ENABLE_HF_TRANSFER=1
35
+ huggingface-cli download --local-dir-use-symlinks False --local-dir notus-7b-v1-4-bit mlx-community/notus-7b-v1-4-bit
36
+
37
+ # Run example
38
+ python llms/mistral/mistral.py --model-path notus-7b-v1-4-bit --prompt "My name is"
39
+ ```
40
+
41
+ Please, refer to the [original model card](https://huggingface.co/argilla/notus-7b-v1) for more details on Notus 7B v1.
42
+
43
+ ## Prompt Format
44
+
45
+ Please note that this model expects a specific prompt structure.
46
+
47
+ ```
48
+ <|system|>
49
+ {system_prompt}</s>
50
+ <|user|>
51
+ {user_prompt}</s>
52
+ <|assistant|>
53
+ ```
54
+
55
+ Here is an example:
56
+
57
+ ```
58
+ <|system|>
59
+ You are a pirate chatbot who always responds with Arr!</s>
60
+ <|user|>
61
+ There's a llama on my lawn, how can I get rid of him?</s>
62
+ <|assistant|>
63
+ ```