Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
tags:
|
6 |
+
- control-vector
|
7 |
+
- creative-writing
|
8 |
+
base_model:
|
9 |
+
- gghfez/WizardLM2-22b-RP
|
10 |
+
---
|
11 |
+
|
12 |
+
## gghfez/WizardLM-2-22B-RP-control-vectors
|
13 |
+
|
14 |
+
Creative Writing control-vectors for [gghfez/WizardLM-2-22B-RP](https://huggingface.co/WizardLM-2-22B-RP)
|
15 |
+
|
16 |
+
## Usage example
|
17 |
+
|
18 |
+
### llama.cpp
|
19 |
+
```sh
|
20 |
+
llama-server --model wizardlm-2-22b-rp-q6_k.gguf [other CLI arguments] \
|
21 |
+
--control-vector WizardLM-2-22B-RP-character_focus__debias.gguf \
|
22 |
+
--control-vector WizardLM-2-22B-RP-character_focus__narration.gguf
|
23 |
+
```
|
24 |
+
|
25 |
+
### Exllamav2
|
26 |
+
|
27 |
+
Clone this repo into the same directory as the exl2 model but with -vectors appended:
|
28 |
+
`huggingface-cli download gghfez/WizardLM-2-22B-RP-control-vectors --local-dir WizardLM-2-22B-RP-exl2-vectors
|
29 |
+
|
30 |
+
So you should have something like
|
31 |
+
/models/WizardLM-2-22B-RP-exl2
|
32 |
+
/models/WizardLM-2-22B-RP-exl2-vectors
|
33 |
+
|
34 |
+
Follow instructions the instructions below for TabbyAPI or Exui:
|
35 |
+
|
36 |
+
#### TabbyAPI Fork
|
37 |
+
|
38 |
+
1. `git clone https://github.com/gapeleon/tabbyAPI-control-vectors`
|
39 |
+
2. `cd tabbyAPI-control-vectors`
|
40 |
+
3. ./start.sh #let it build the venv, etc then exit
|
41 |
+
4. `source venv/bin/activate` #to enter the venv
|
42 |
+
5. `pip install gguf` #to read the vectors
|
43 |
+
6. `cp config_sample_control-vectors.yml config.yml`
|
44 |
+
7. set the model_dir and model_name in config.yml
|
45 |
+
8. tweak the control-vectors how you want them
|
46 |
+
9. ./start.sh
|
47 |
+
|
48 |
+
#### Exui Fork
|
49 |
+
|
50 |
+
|
51 |
+
1. `git clone https://github.com/gapeleon/exui-control-vectors`
|
52 |
+
2. Set it up / install it the same as regular exui
|
53 |
+
3. `pip install gguf` if you don't have it already
|
54 |
+
4. Start exui `python server.py`
|
55 |
+
5. Configure the model as usual.
|
56 |
+
6. Tick "Enable control vectors" and wait for them to appear in the UI
|
57 |
+
7. Enable the vectors + direction + scaling you want, then hit "Load Model"
|
58 |
+
|
59 |
+
|
60 |
+
Based off Version v3.0 of [creative-writing-control-vectors-v3.0](https://huggingface.co/jukofyork/creative-writing-control-vectors-v3.0/).
|
61 |
+
|
62 |
+
See here for more info about control-vectors [creative-writing-control-vectors-v3.0](https://huggingface.co/jukofyork/creative-writing-control-vectors-v3.0/)
|