File size: 1,432 Bytes
ba37c8f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model:
- powermove72/Shark-1
- S-miguel/The-Trinity-Coder-7B
tags:
- merge
- mergekit
- lazymergekit
- powermove72/Shark-1
- S-miguel/The-Trinity-Coder-7B
---

# Shark-Coder

Shark-Coder is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [powermove72/Shark-1](https://huggingface.co/powermove72/Shark-1)
* [S-miguel/The-Trinity-Coder-7B](https://huggingface.co/S-miguel/The-Trinity-Coder-7B)

## 🧩 Configuration

```yaml
slices:
  - sources:
    - model: powermove72/Shark-1
      layer_range: [0, 16]
  - sources:
    - model: S-miguel/The-Trinity-Coder-7B
      layer_range: [16, 32]
merge_method: passthrough
tokenizer_source: union
dtype: float16
    ```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "powermove72/Shark-Coder"
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.float16,
    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"])
```