File size: 1,233 Bytes
8bfd3a6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45d2cb3
 
 
8bfd3a6
45d2cb3
8bfd3a6
45d2cb3
8bfd3a6
45d2cb3
 
8bfd3a6
45d2cb3
 
8bfd3a6
 
45d2cb3
8bfd3a6
45d2cb3
 
 
8bfd3a6
 
45d2cb3
8bfd3a6
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
---
tags:
- autotrain
- text-generation-inference
- text-generation
- peft
library_name: transformers
widget:
  - messages:
      - role: user
        content: What is your favorite condiment?
license: other
---

# Model Trained Using AutoTrain

This model was trained using AutoTrain. For more information, please visit [AutoTrain](https://hf.co/docs/autotrain).

# Usage

```bash
pip install peft
```

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import AutoPeftModelForCausalLM, PeftConfig

model_id = "Aryan-401/phi-3-mini-4k-instruct-finetune-guanaco"
peft_model=AutoPeftModelForCausalLM.from_pretrained(model_id)

model = peft_model.merge_and_unload()
tokenizer = AutoTokenizer.from_pretrained(model_id)

messages = [
    {"role": "user", "content": "What is the Value of Pi?"}
]
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

model = model.to(device).eval()

input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
output_ids = model.generate(input_ids.to(device), max_length= 1000)
response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)

print(response)
```