README.md

---
license: mit
tags:
- moe
- frankenmoe
- merge
- mergekit
- lazymergekit
- Open-Orca/Mistral-7B-OpenOrca
- Crystalcareai/Evol-Mistral
base_model:
- Open-Orca/Mistral-7B-OpenOrca
- Crystalcareai/Evol-Mistral
---

# Evolorxa-14b

Evolorxa-14b is a Mixure of Experts (MoE) made with the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [Open-Orca/Mistral-7B-OpenOrca](https://huggingface.co/Open-Orca/Mistral-7B-OpenOrca)
* [Crystalcareai/Evol-Mistral](https://huggingface.co/Crystalcareai/Evol-Mistral)

## 🧩 Configuration

```yaml
slices:
  - sources:
      - model: Open-Orca/Mistral-7B-OpenOrca
        layer_range: [0, 32]
      - model: Crystalcareai/Evol-Mistral
        layer_range: [0, 32]
merge_method: slerp
base_model: Open-Orca/Mistral-7B-OpenOrca
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16
experts:
  - source_model: Open-Orca/Mistral-7B-OpenOrca
    positive_prompts:
    - "chat"
    - "reasoning"
    - "Why would"
    - "explain"
  - source_model: Crystalcareai/Evol-Mistral
    positive_prompts:
    - "instruction"
    - "create a"
    - "You must"
    - "Your job"
```

## 💻 Usage

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

from transformers import AutoTokenizer
import transformers
import torch

model = "Crystalcareai/Evolorxa-14b"

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"])
```