Ruqiya's picture
Update README.md
6ec2611 verified
|
raw
history blame
1.42 kB
---
tags:
- merge
- mergekit
- lazymergekit
- gemma-2b
pipeline_tag: text-generation
---
# Merge-Gemma-2b-Models
Merge-Gemma-2b-Models is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [unsloth/gemma-2b](https://huggingface.co/unsloth/gemma-2b)
* [anakin87/gemma-2b-orpo](https://huggingface.co/anakin87/gemma-2b-orpo)
## 🧩 Configuration
```yaml
models:
- model: anakin87/gemma-2b-orpo
# No parameters necessary for base model
- model: unsloth/gemma-2b
parameters:
density: 0.53
weight: 0.45
merge_method: dare_ties
base_model: anakin87/gemma-2b-orpo
parameters:
int8_mask: true
dtype: bfloat16
random_seed: 0
```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Ruqiya/Merge-Gemma-2b-Models"
messages = [{"role": "user", "content": "What is artificial intelligence?"}]
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"])
```