Commit
•
87598cf
1
Parent(s):
2a8999d
Added additional vision example
Browse files
README.md
CHANGED
|
@@ -32,7 +32,11 @@ First in a series of experiments centering around fine-tuning for image captioni
|
|
| 32 |
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
|
| 33 |
|
| 34 |
<h1>Vision Results</h1>
|
|
|
|
|
|
|
| 35 |
<img src="https://i.imgur.com/E9mS4Xb.jpeg" width="400"/>
|
|
|
|
|
|
|
| 36 |
|
| 37 |
* Experiment yielded model that provides shorter, less verbose output for questions about pictures
|
| 38 |
* The likelihood of hallucinations in output has decreased, however, the model can still be easily influenced to be inaccurate by the user
|
|
|
|
| 32 |
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
|
| 33 |
|
| 34 |
<h1>Vision Results</h1>
|
| 35 |
+
|
| 36 |
+
Example 1
|
| 37 |
<img src="https://i.imgur.com/E9mS4Xb.jpeg" width="400"/>
|
| 38 |
+
Example 2
|
| 39 |
+
<img src="https://i.imgur.com/SmTz1Yd.jpeg" width="400"/>
|
| 40 |
|
| 41 |
* Experiment yielded model that provides shorter, less verbose output for questions about pictures
|
| 42 |
* The likelihood of hallucinations in output has decreased, however, the model can still be easily influenced to be inaccurate by the user
|