Update README.md
Browse files
README.md
CHANGED
|
@@ -28,6 +28,8 @@ This is an instruction fine-tuned llama-2 model, using synthetic data generated
|
|
| 28 |
- these models just produce text, what you do with that text is your resonsibility
|
| 29 |
- many people and industries deal with "sensitive" content; imagine if a court stenographer's eqipment filtered illegal content - it would be useless
|
| 30 |
|
|
|
|
|
|
|
| 31 |
### Prompt format
|
| 32 |
|
| 33 |
The training code was updated to randomize newline vs space:
|
|
|
|
| 28 |
- these models just produce text, what you do with that text is your resonsibility
|
| 29 |
- many people and industries deal with "sensitive" content; imagine if a court stenographer's eqipment filtered illegal content - it would be useless
|
| 30 |
|
| 31 |
+
Huge thank you to the folks over at [a16z](https://a16z.com/) for sponsoring the costs associated with building models and associated tools!
|
| 32 |
+
|
| 33 |
### Prompt format
|
| 34 |
|
| 35 |
The training code was updated to randomize newline vs space:
|