File size: 1,499 Bytes
811cf49
0ed1268
 
 
811cf49
 
 
 
0ed1268
 
811cf49
 
 
 
 
0ed1268
 
811cf49
 
 
 
 
 
8dcf265
0ed1268
 
 
 
811cf49
0ed1268
 
 
 
811cf49
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model:
- Sagicc/Occam-2B-CCoT
- allenai/tmax-2b
tags:
- merge
- mergekit
- lazymergekit
- Sagicc/Occam-2B-CCoT
- allenai/tmax-2b
---

# Papercut-2B

Papercut-2B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [Sagicc/Occam-2B-CCoT](https://huggingface.co/Sagicc/Occam-2B-CCoT)
* [allenai/tmax-2b](https://huggingface.co/allenai/tmax-2b)

## 🧩 Configuration

```yaml
models:
  - model: Qwen/Qwen3.5-2B # Base
  - model: Sagicc/Occam-2B-CCoT # Short Thinking + Tool Calling
    parameters:
      weight: 1.0
      density: 0.9
      gamma: 0.01
  - model: allenai/tmax-2b # Terminal Agent
    parameters:
      weight: 1.0
      density: 0.9
      gamma: 0.01
merge_method: breadcrumbs_ties
base_model: Qwen/Qwen3.5-2B

```

## 💻 Usage

```python
!uv pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "sillykiwi/Papercut-2B"
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"])
```