---
tags:
- merge
- mergekit
- lazymergekit
- Eric111/Mayo
- NousResearch/Hermes-2-Pro-Mistral-7B
- mistralai/Mistral-7B-Instruct-v0.2
- NousResearch/Yarn-Mistral-7b-128k
- Kukedlc/MyModelsMerge-7b
base_model:
- Eric111/Mayo
- NousResearch/Hermes-2-Pro-Mistral-7B
- mistralai/Mistral-7B-Instruct-v0.2
- NousResearch/Yarn-Mistral-7b-128k
- Kukedlc/MyModelsMerge-7b
license: apache-2.0
---

# NeuralContext-7b

NeuralContext-7b is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [Eric111/Mayo](https://huggingface.co./Eric111/Mayo)
* [NousResearch/Hermes-2-Pro-Mistral-7B](https://huggingface.co./NousResearch/Hermes-2-Pro-Mistral-7B)
* [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co./mistralai/Mistral-7B-Instruct-v0.2)
* [NousResearch/Yarn-Mistral-7b-128k](https://huggingface.co./NousResearch/Yarn-Mistral-7b-128k)
* [Kukedlc/MyModelsMerge-7b](https://huggingface.co./Kukedlc/MyModelsMerge-7b)

## 🧩 Configuration

```yaml
models:
  - model: NousResearch/Yarn-Mistral-7b-128k
    # No parameters necessary for base model
  - model: Eric111/Mayo
    parameters:
      density: 0.33
      weight: 0.2
  - model: NousResearch/Hermes-2-Pro-Mistral-7B
    parameters:
      density: 0.66
      weight: 0.2
  - model: mistralai/Mistral-7B-Instruct-v0.2
    parameters:
      density: 0.66
      weight: 0.2
  - model: NousResearch/Yarn-Mistral-7b-128k
    parameters:
      density: 0.66
      weight: 0.2
  - model: Kukedlc/MyModelsMerge-7b
    parameters:
      density: 0.55
      weight: 0.2
merge_method: dare_ties
base_model: NousResearch/Yarn-Mistral-7b-128k
parameters:
  int8_mask: true
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Kukedlc/NeuralContext-7b-v1"
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"])
```