Commit
•
5cdd4d8
1
Parent(s):
b35fab6
Update README.md
Browse filesAdded additional details and sample question
README.md
CHANGED
@@ -17,11 +17,18 @@ datasets:
|
|
17 |
An initial foray into the world of fine-tuning. The goal of this release was to amplify the quality of the original model's responses, in particular for vision use cases*
|
18 |
|
19 |
|
20 |
-
|
21 |
* [Excalibur-7b](https://huggingface.co/InferenceIllusionist/Excalibur-7b) fine-tuned with Direct Preference Optimization (DPO) using Intel/orca_dpo_pairs
|
22 |
* This is a quick experiment to determine the impact of DPO finetuning on the original base model
|
23 |
* Ran for a little over an hour on a single A100
|
24 |
* Internal benchmarks showed improvement over base model, awaiting final results
|
25 |
* Precision: bfloat16
|
26 |
|
27 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
An initial foray into the world of fine-tuning. The goal of this release was to amplify the quality of the original model's responses, in particular for vision use cases*
|
18 |
|
19 |
|
20 |
+
## Notes & Methodology
|
21 |
* [Excalibur-7b](https://huggingface.co/InferenceIllusionist/Excalibur-7b) fine-tuned with Direct Preference Optimization (DPO) using Intel/orca_dpo_pairs
|
22 |
* This is a quick experiment to determine the impact of DPO finetuning on the original base model
|
23 |
* Ran for a little over an hour on a single A100
|
24 |
* Internal benchmarks showed improvement over base model, awaiting final results
|
25 |
* Precision: bfloat16
|
26 |
|
27 |
+
|
28 |
+
## Sample Question - Vision
|
29 |
+
<img src="https://i.imgur.com/7aRWtzU.jpeg" width="425"/>
|
30 |
+
|
31 |
+
<b>Requires additional [mistral-7b-mmproj-v1.5-Q4_1.gguf](https://huggingface.co/koboldcpp/mmproj/tree/main) file for vision functionality</b>
|
32 |
+
|
33 |
+
Select up the gguf file of your choice in Kobold as usual, then make sure to choose the mmproj file above in the LLaVA mmproj field of the model submenu:
|
34 |
+
<img src="https://i.imgur.com/u1vo8Rs.jpeg" width="550"/>
|