File size: 1,656 Bytes
3a228ce
 
2941a68
3a228ce
2941a68
3a228ce
 
 
2941a68
 
3a228ce
 
2941a68
3a228ce
2941a68
 
3a228ce
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2941a68
3a228ce
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model:
- recoilme/recoilme-gemma-2-9B-v0.4
- BAAI/Gemma2-9B-IT-Simpo-Infinity-Preference
library_name: transformers
tags:
- merge
- mergekit
language:
- en
---

# Gemma 2 Lyco Infinity 9B

Gemma 2 Lyco Infinity 9B is a merge of the following models using [Mergekit](https://github.com/arcee-ai/mergekit):
* [recoilme/recoilme-gemma-2-9B-v0.4](https://huggingface.co/recoilme/recoilme-gemma-2-9B-v0.4)
* [BAAI/Gemma2-9B-IT-Simpo-Infinity-Preference](https://huggingface.co/BAAI/Gemma2-9B-IT-Simpo-Infinity-Preference)

## 🧩 Configuration

```yaml
base_model: recoilme/recoilme-gemma-2-9B-v0.4
models:
  - model: recoilme/recoilme-gemma-2-9B-v0.4
    # No parameters necessary for base model
  - model: BAAI/Gemma2-9B-IT-Simpo-Infinity-Preference
    parameters:
      density: 0.525
      weight: 1
merge_method: dare_ties
tokenizer_source: base
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "AELLM/gemma-2-lyco-infinity-9b"
messages = [{"role": "user", "content": "You're an AI assistant for a time traveler. How do you help them blend into different historical eras without causing any time paradoxes?"}]

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"])
```