---
license: apache-2.0
tags:
- moe
- frankenmoe
- merge
- mergekit
- lazymergekit
- bardsai/jaskier-7b-dpo-v6.1
- CultriX/NeuralTrix-7B-dpo
base_model:
- bardsai/jaskier-7b-dpo-v6.1
- CultriX/NeuralTrix-7B-dpo
---

# megatron_2.1_MoE_2x7B

megatron_2.1_MoE_2x7B is a Mixure of Experts (MoE) made with the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [bardsai/jaskier-7b-dpo-v6.1](https://huggingface.co/bardsai/jaskier-7b-dpo-v6.1)
* [CultriX/NeuralTrix-7B-dpo](https://huggingface.co/CultriX/NeuralTrix-7B-dpo)

## 🧩 Configuration

```yaml
base_model: bardsai/jaskier-7b-dpo-v6.1
gate_mode: hidden
dtype: bfloat16
experts:
  - source_model: bardsai/jaskier-7b-dpo-v6.1
    positive_prompts:
    - "Mathematics"
    - "Physics"
    - "Chemistry"
    - "Biology"
    - "Medicine"
    - "Engineering"
    - "Computer Science"

    negative_prompts:
    - "Earth Sciences (Geology, Meteorology, Oceanography)"
    - "Environmental Science"
    - "Astronomy and Space Science"
    - "Psychology"
    - "Sociology"
    - "Anthropology"
    - "Political Science"
    - "Economics"
    - "Education"
    - "Law"
    - "Theology and Religious Studies"
    - "Communication Studies"
    - "Business and Management"
    - "Agricultural Sciences"
    - "Nutrition and Food Science"
    - "Sports Science"

  - source_model: CultriX/NeuralTrix-7B-dpo
    positive_prompts:
    - "Earth Sciences (Geology, Meteorology, Oceanography)"
    - "Environmental Science"
    - "Astronomy and Space Science"
    - "Psychology"
    - "Sociology"
    - "Anthropology"
    - "Political Science"
    - "Economics"
    - "Education"
    - "Law"
    - "Theology and Religious Studies"
    - "Communication Studies"
    - "Business and Management"
    - "Agricultural Sciences"
    - "Nutrition and Food Science"
    - "Sports Science"

    negative_prompts:
    - "Mathematics"
    - "Physics"
    - "Chemistry"
    - "Biology"
    - "Medicine"
    - "Engineering"
    - "Computer Science"
```

## 💻 Usage

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

from transformers import AutoTokenizer
import transformers
import torch

model = "Eurdem/megatron_2.1_MoE_2x7B"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)

messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
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"])
```