Commit
•
469d285
1
Parent(s):
620ea14
Update README.md
Browse files
README.md
CHANGED
@@ -28,6 +28,23 @@ Release Notes:
|
|
28 |
|
29 |
Mistral-7b-02 base model was fine-tuned using the [RealWorldQA dataset](https://huggingface.co/datasets/visheratin/realworldqa), originally provided by the X.Ai Team here: https://x.ai/blog/grok-1.5v
|
30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
- **Developed by:** InferenceIllusionist
|
32 |
- **License:** apache-2.0
|
33 |
- **Finetuned from model :** mistral-community/Mistral-7B-v0.2
|
|
|
28 |
|
29 |
Mistral-7b-02 base model was fine-tuned using the [RealWorldQA dataset](https://huggingface.co/datasets/visheratin/realworldqa), originally provided by the X.Ai Team here: https://x.ai/blog/grok-1.5v
|
30 |
|
31 |
+
<b>Initial Results</b>
|
32 |
+
<img src="https://i.imgur.com/E9mS4Xb.jpeg" width="400"/>
|
33 |
+
* Experiment yielded model that provides shorter, less verbose output for questions about pictures
|
34 |
+
* The likelihood of hallucinations in output has decreased, however, the model can still be easily influenced to be inaccurate by the user
|
35 |
+
* Best suited for captioning use cases that require concise descriptions and low token counts
|
36 |
+
* This model lacks the conversational prose of Excalibur-7b-DPO and is much "drier" in tone
|
37 |
+
|
38 |
+
*<b>Requires additional mmproj file. You have two options for vision functionality (available inside this repo):</b>
|
39 |
+
* [Quantized - Limited VRAM Option (197mb)](https://huggingface.co/InferenceIllusionist/Excalibur-7b-DPO-GGUF/resolve/main/mistral-7b-mmproj-v1.5-Q4_1.gguf?download=true)
|
40 |
+
* [Unquantized - Premium Option / Best Quality (596mb)](https://huggingface.co/InferenceIllusionist/Excalibur-7b-DPO-GGUF/resolve/main/mmproj-model-f16.gguf?download=true)
|
41 |
+
|
42 |
+
Select the gguf file of your choice in [Koboldcpp](https://github.com/LostRuins/koboldcpp/releases/) as usual, then make sure to choose the mmproj file above in the LLaVA mmproj field of the model submenu:
|
43 |
+
<img src="https://i.imgur.com/x8vqH29.png" width="425"/>
|
44 |
+
|
45 |
+
## Prompt Format
|
46 |
+
Use Alpaca for best results.
|
47 |
+
|
48 |
- **Developed by:** InferenceIllusionist
|
49 |
- **License:** apache-2.0
|
50 |
- **Finetuned from model :** mistral-community/Mistral-7B-v0.2
|