File size: 1,930 Bytes
7723e9c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
---
tags:
- merge
- mergekit
- lazymergekit
- RaduGabriel/MUZD
- RaduGabriel/Mistral-Instruct-Ukrainian-SFT
- Radu1999/MisterUkrainianDPO
- CultriX/NeuralTrix-7B-dpo
base_model:
- RaduGabriel/MUZD
- RaduGabriel/Mistral-Instruct-Ukrainian-SFT
- Radu1999/MisterUkrainianDPO
- CultriX/NeuralTrix-7B-dpo
---

# NeuralPipe-7B-slerp

NeuralPipe-7B-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [RaduGabriel/MUZD](https://huggingface.co/RaduGabriel/MUZD)
* [RaduGabriel/Mistral-Instruct-Ukrainian-SFT](https://huggingface.co/RaduGabriel/Mistral-Instruct-Ukrainian-SFT)
* [Radu1999/MisterUkrainianDPO](https://huggingface.co/Radu1999/MisterUkrainianDPO)
* [CultriX/NeuralTrix-7B-dpo](https://huggingface.co/CultriX/NeuralTrix-7B-dpo)

## 🧩 Configuration

```yaml
models:
  - model: RaduGabriel/MUZD
    parameters:
      weight: 0.3
  - model: RaduGabriel/Mistral-Instruct-Ukrainian-SFT
    parameters:
      weight: 0.3
  - model: Radu1999/MisterUkrainianDPO
    parameters:
      weight: 0.1
  - model: CultriX/NeuralTrix-7B-dpo
    parameters:
      weight: 0.3

merge_method: task_arithmetic
base_model: mistralai/Mistral-7B-v0.1
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "RaduGabriel/NeuralPipe-7B-slerp"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```