Commit
•
1ec0ea0
1
Parent(s):
96f8ee3
Update README.md
Browse files
README.md
CHANGED
@@ -7,7 +7,7 @@ language:
|
|
7 |
- en
|
8 |
---
|
9 |
|
10 |
-
![An eagle
|
11 |
|
12 |
### Huggingface RWKV EagleX 7B v2 Model
|
13 |
|
@@ -19,6 +19,13 @@ language:
|
|
19 |
>
|
20 |
> This is not an instruct tune model! (soon...)
|
21 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
## Evaluation
|
23 |
|
24 |
The following shows the progression of the model from 1.1T trained to 2.25T trained.
|
|
|
7 |
- en
|
8 |
---
|
9 |
|
10 |
+
![An eagle flying high up in the sky](https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F304f2c7a-fc67-4df4-ba57-c6f38f86826c_2688x1536.png)
|
11 |
|
12 |
### Huggingface RWKV EagleX 7B v2 Model
|
13 |
|
|
|
19 |
>
|
20 |
> This is not an instruct tune model! (soon...)
|
21 |
|
22 |
+
## Quickstart with the hugging face transformer library
|
23 |
+
|
24 |
+
```
|
25 |
+
model = AutoModelForCausalLM.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True).to(torch.float32)
|
26 |
+
tokenizer = AutoTokenizer.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True)
|
27 |
+
```
|
28 |
+
|
29 |
## Evaluation
|
30 |
|
31 |
The following shows the progression of the model from 1.1T trained to 2.25T trained.
|