pankajmathur
commited on
Commit
β’
0f8a1ed
1
Parent(s):
6773474
Update README.md
Browse files
README.md
CHANGED
@@ -1,16 +1,21 @@
|
|
1 |
---
|
2 |
license: llama3
|
|
|
|
|
|
|
|
|
3 |
---
|
4 |
-
Introducing
|
5 |
|
6 |
-
This
|
7 |
|
|
|
8 |
|
|
|
9 |
|
10 |
-
|
11 |
|
12 |
-
|
13 |
-
ChatML
|
14 |
|
15 |
```
|
16 |
<|im_start|>system
|
@@ -21,7 +26,7 @@ ChatML
|
|
21 |
{asistant}<|im_end|>
|
22 |
```
|
23 |
|
24 |
-
|
25 |
```
|
26 |
messages = [
|
27 |
{"role": "system", "content": "You are helpful AI asistant."},
|
|
|
1 |
---
|
2 |
license: llama3
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
library_name: transformers
|
6 |
+
pipeline_tag: text2text-generation
|
7 |
---
|
8 |
+
π **Introducing BrokenLlama-3-8b: 100% Full Finetuning No DPO added Enjoy!** π
|
9 |
|
10 |
+
This bad boy is a fully fine-tuned version of the already awesome Meta-Llama-3-8B, but we've cranked it up to 11 by attempting to remove alignment and biases using a super special curated dataset π with 8192 sequence length.
|
11 |
|
12 |
+
BrokenLlama-3-8b went through a crazy 48-hour training session on some seriously powerful hardware, so you know it's ready to rock your world. πͺ
|
13 |
|
14 |
+
With skills that'll blow your mind, BrokenLlama-3-8b can chat, code, and even do some fancy function calls. π€
|
15 |
|
16 |
+
But watch out! This llama is a wild one and will do pretty much anything you ask, even if it's a bit naughty. π Make sure to keep it in check with your own alignment layer before letting it loose in the wild.
|
17 |
|
18 |
+
To get started with this incredible model, just use the ChatML prompt template and let the magic happen. It's so easy, even a llama could do it! π¦
|
|
|
19 |
|
20 |
```
|
21 |
<|im_start|>system
|
|
|
26 |
{asistant}<|im_end|>
|
27 |
```
|
28 |
|
29 |
+
ChatML prompt template is available as a chat template, which means you can format messages using the tokenizer.apply_chat_template() method:
|
30 |
```
|
31 |
messages = [
|
32 |
{"role": "system", "content": "You are helpful AI asistant."},
|