change name
Browse files
README.md
CHANGED
@@ -12,8 +12,10 @@ language:
|
|
12 |
|
13 |
**The Faro chat model focuses on practicality and long-context modeling. It handles various downstream tasks with higher quality, delivering stable and reliable results even when inputs contain lengthy documents or complex instructions. Faro seamlessly works in both English and Chinese.**
|
14 |
|
15 |
-
# Faro-Yi-34B
|
16 |
-
Faro-Yi-34B
|
|
|
|
|
17 |
|
18 |
## How to Use
|
19 |
|
@@ -25,7 +27,7 @@ import requests
|
|
25 |
from PyPDF2 import PdfReader
|
26 |
from vllm import LLM, SamplingParams
|
27 |
|
28 |
-
llm = LLM(model="wenbopan/Faro-Yi-34B
|
29 |
|
30 |
pdf_data = io.BytesIO(requests.get("https://arxiv.org/pdf/2303.08774.pdf").content)
|
31 |
document = "".join(page.extract_text() for page in PdfReader(pdf_data).pages) # 100 pages
|
@@ -44,8 +46,8 @@ print(output[0].outputs[0].text)
|
|
44 |
```python
|
45 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
46 |
|
47 |
-
model = AutoModelForCausalLM.from_pretrained('wenbopan/Faro-Yi-34B
|
48 |
-
tokenizer = AutoTokenizer.from_pretrained('wenbopan/Faro-Yi-34B
|
49 |
messages = [
|
50 |
{"role": "system", "content": "You are a helpful assistant. Always answer with a short response."},
|
51 |
{"role": "user", "content": "Tell me what is Pythagorean theorem like you are a pirate."}
|
@@ -57,4 +59,4 @@ response = tokenizer.decode(generated_ids[0], skip_special_tokens=True) # Aye, m
|
|
57 |
|
58 |
</details>
|
59 |
|
60 |
-
For more info please refer to [wenbopan/Faro-Yi-9B
|
|
|
12 |
|
13 |
**The Faro chat model focuses on practicality and long-context modeling. It handles various downstream tasks with higher quality, delivering stable and reliable results even when inputs contain lengthy documents or complex instructions. Faro seamlessly works in both English and Chinese.**
|
14 |
|
15 |
+
# Faro-Yi-34B
|
16 |
+
Faro-Yi-34B is an improved [Yi-34B-200K](https://huggingface.co/01-ai/Yi-34B-200K) with extensive instruction tuning on [Fusang-V1](https://huggingface.co/datasets/wenbopan/Fusang-v1). Compared to Yi-34B-200K, Faro-Yi-34B has gained greater capability in various downstream tasks and long-context modeling thanks to the large-scale synthetic data in Fusang-V1.
|
17 |
+
|
18 |
+
Just like Yi-34B-200K, Faro-Yi-34B supports up to 200K context length.
|
19 |
|
20 |
## How to Use
|
21 |
|
|
|
27 |
from PyPDF2 import PdfReader
|
28 |
from vllm import LLM, SamplingParams
|
29 |
|
30 |
+
llm = LLM(model="wenbopan/Faro-Yi-34B")
|
31 |
|
32 |
pdf_data = io.BytesIO(requests.get("https://arxiv.org/pdf/2303.08774.pdf").content)
|
33 |
document = "".join(page.extract_text() for page in PdfReader(pdf_data).pages) # 100 pages
|
|
|
46 |
```python
|
47 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
48 |
|
49 |
+
model = AutoModelForCausalLM.from_pretrained('wenbopan/Faro-Yi-34B', device_map="cuda")
|
50 |
+
tokenizer = AutoTokenizer.from_pretrained('wenbopan/Faro-Yi-34B')
|
51 |
messages = [
|
52 |
{"role": "system", "content": "You are a helpful assistant. Always answer with a short response."},
|
53 |
{"role": "user", "content": "Tell me what is Pythagorean theorem like you are a pirate."}
|
|
|
59 |
|
60 |
</details>
|
61 |
|
62 |
+
For more info please refer to [wenbopan/Faro-Yi-9B](https://huggingface.co/wenbopan/Faro-Yi-9B)
|