Commit
·
958e42a
1
Parent(s):
85869ab
Update README.md
Browse files
README.md
CHANGED
|
@@ -58,7 +58,6 @@ pinned: false
|
|
| 58 |
|
| 59 |
|
| 60 |
<hr>
|
| 61 |
-
|
| 62 |
<p align="center" style="margin-top: 0px; font-size: 1.5em; background-color: #3c72db; padding: 0.5em; border-radius: 0.5em; color: white; font-weight: bold;">
|
| 63 |
<a href="https://huggingface.co/openchat/openchat_3.5" style="text-decoration: none; color: white;">
|
| 64 |
<span style="font-size: 1.6em; font-family: 'Helvetica'; letter-spacing: 0.2em">OPENCHAT</span>
|
|
@@ -74,9 +73,7 @@ pinned: false
|
|
| 74 |
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat_grok.png" style="width: 47%;">
|
| 75 |
</div>
|
| 76 |
</p>
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
|
| 81 |
- OpenChat is an innovative library of **open-source language models**, fine-tuned with [**C-RLFT**](https://arxiv.org/pdf/2309.11235.pdf) - a strategy inspired by offline reinforcement learning.
|
| 82 |
- Our models learn from mixed-quality data without preference labels, delivering exceptional performance on par with `ChatGPT`, even with a `7B` model which can be run on a **consumer GPU (e.g. RTX 3090)**.
|
|
|
|
| 58 |
|
| 59 |
|
| 60 |
<hr>
|
|
|
|
| 61 |
<p align="center" style="margin-top: 0px; font-size: 1.5em; background-color: #3c72db; padding: 0.5em; border-radius: 0.5em; color: white; font-weight: bold;">
|
| 62 |
<a href="https://huggingface.co/openchat/openchat_3.5" style="text-decoration: none; color: white;">
|
| 63 |
<span style="font-size: 1.6em; font-family: 'Helvetica'; letter-spacing: 0.2em">OPENCHAT</span>
|
|
|
|
| 73 |
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat_grok.png" style="width: 47%;">
|
| 74 |
</div>
|
| 75 |
</p>
|
| 76 |
+
<hr>
|
|
|
|
|
|
|
| 77 |
|
| 78 |
- OpenChat is an innovative library of **open-source language models**, fine-tuned with [**C-RLFT**](https://arxiv.org/pdf/2309.11235.pdf) - a strategy inspired by offline reinforcement learning.
|
| 79 |
- Our models learn from mixed-quality data without preference labels, delivering exceptional performance on par with `ChatGPT`, even with a `7B` model which can be run on a **consumer GPU (e.g. RTX 3090)**.
|