Commit
β’
7c68121
1
Parent(s):
dc966c6
Update README.md
Browse files
README.md
CHANGED
@@ -27,13 +27,16 @@ inference: false
|
|
27 |
<em>An open and efficient LLM for Dutch</em>
|
28 |
</div>
|
29 |
|
30 |
-
<blockquote class="tip">
|
31 |
<p align="center">
|
32 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b">π±ββοΈ Base version</a> (this one) -
|
33 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-instruct">π€ Instruct version</a> -
|
34 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-chat">π¬ Chat version</a> -
|
35 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-GGUF">π GGUF of base model</a>
|
36 |
</p>
|
|
|
|
|
|
|
37 |
</blockquote>
|
38 |
|
39 |
Fietje is an adapated version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2), tailored to Dutch text generation by training on 28B tokens. It is small and efficient with a size of 2.7 billion parameters while performing almost on par with more powerful Dutch LLMs of twice its size like [GEITje 7B Ultra](https://huggingface.co/BramVanroy/GEITje-7B-ultra).
|
|
|
27 |
<em>An open and efficient LLM for Dutch</em>
|
28 |
</div>
|
29 |
|
30 |
+
<blockquote class="tip" style="padding: 1.5em; border: 0">
|
31 |
<p align="center">
|
32 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b">π±ββοΈ Base version</a> (this one) -
|
33 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-instruct">π€ Instruct version</a> -
|
34 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-chat">π¬ Chat version</a> -
|
35 |
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-GGUF">π GGUF of base model</a>
|
36 |
</p>
|
37 |
+
<p align="center" style="text-align: center; margin: 0">
|
38 |
+
<a href="https://huggingface.co/spaces/BramVanroy/fietje-2b"><strong>Chat with Fietje here!</strong></a>
|
39 |
+
</p>
|
40 |
</blockquote>
|
41 |
|
42 |
Fietje is an adapated version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2), tailored to Dutch text generation by training on 28B tokens. It is small and efficient with a size of 2.7 billion parameters while performing almost on par with more powerful Dutch LLMs of twice its size like [GEITje 7B Ultra](https://huggingface.co/BramVanroy/GEITje-7B-ultra).
|