README.md

---
tags:
- merge
- mergekit
- lazymergekit
- mistralai/Mistral-7B-Instruct-v0.3
base_model:
- mistralai/Mistral-7B-Instruct-v0.3
license: apache-2.0
---

# AI-M3-10.7Bv2

AI-M3-10.7Bv2 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3)

## 🧩 Configuration

```yaml
dtype: float16
base_model: mistralai/Mistral-7B-Instruct-v0.3
merge_method: task_arithmetic
slices:
- sources:
    - model: mistralai/Mistral-7B-Instruct-v0.3
      layer_range: [0, 26]
- sources:
    - model: mistralai/Mistral-7B-Instruct-v0.3
      layer_range: [10, 32]
parameters:
  t: 1
  weight: 1.0
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "sydonayrex/AI-M3-10.7Bv2"
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"])
```