Upload README.md
Browse files
README.md
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
pipeline_tag: text-generation
|
6 |
+
inference: false
|
7 |
+
tags:
|
8 |
+
- transformers
|
9 |
+
- gguf
|
10 |
+
- imatrix
|
11 |
+
- SOLAR-10.7B-Instruct-v1.0
|
12 |
+
---
|
13 |
+
Quantizations of https://huggingface.co/upstage/SOLAR-10.7B-Instruct-v1.0
|
14 |
+
|
15 |
+
# From original readme
|
16 |
+
|
17 |
+
# **Usage Instructions**
|
18 |
+
|
19 |
+
This model has been fine-tuned primarily for single-turn conversation, making it less suitable for multi-turn conversations such as chat.
|
20 |
+
|
21 |
+
### **Version**
|
22 |
+
|
23 |
+
Make sure you have the correct version of the transformers library installed:
|
24 |
+
|
25 |
+
```sh
|
26 |
+
pip install transformers==4.35.2
|
27 |
+
```
|
28 |
+
|
29 |
+
### **Loading the Model**
|
30 |
+
|
31 |
+
Use the following Python code to load the model:
|
32 |
+
|
33 |
+
```python
|
34 |
+
import torch
|
35 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
36 |
+
|
37 |
+
tokenizer = AutoTokenizer.from_pretrained("Upstage/SOLAR-10.7B-Instruct-v1.0")
|
38 |
+
model = AutoModelForCausalLM.from_pretrained(
|
39 |
+
"Upstage/SOLAR-10.7B-Instruct-v1.0",
|
40 |
+
device_map="auto",
|
41 |
+
torch_dtype=torch.float16,
|
42 |
+
)
|
43 |
+
```
|
44 |
+
|
45 |
+
### **Conducting Single-Turn Conversation**
|
46 |
+
|
47 |
+
```python
|
48 |
+
conversation = [ {'role': 'user', 'content': 'Hello?'} ]
|
49 |
+
|
50 |
+
prompt = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
|
51 |
+
|
52 |
+
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
|
53 |
+
outputs = model.generate(**inputs, use_cache=True, max_length=4096)
|
54 |
+
output_text = tokenizer.decode(outputs[0])
|
55 |
+
print(output_text)
|
56 |
+
```
|
57 |
+
|
58 |
+
Below is an example of the output.
|
59 |
+
```
|
60 |
+
<s> ### User:
|
61 |
+
Hello?
|
62 |
+
|
63 |
+
### Assistant:
|
64 |
+
Hello, how can I assist you today? Please feel free to ask any questions or request help with a specific task.</s>
|
65 |
+
```
|