README.md

---
license: apache-2.0
tags:
- moe
- merge
- mergekit
model-index:
- name: Mixolar-4x7b
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 71.08
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=shadowml/Mixolar-4x7b
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 88.44
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=shadowml/Mixolar-4x7b
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 66.29
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=shadowml/Mixolar-4x7b
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 71.81
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=shadowml/Mixolar-4x7b
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 83.58
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=shadowml/Mixolar-4x7b
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 63.91
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=shadowml/Mixolar-4x7b
      name: Open LLM Leaderboard
---

# Mixolar-4x7b

This model is a Mixure of Experts (MoE) made with [mergekit](https://github.com/cg123/mergekit) (mixtral branch). It uses the following base models:
 * [kyujinpy/Sakura-SOLAR-Instruct](https://huggingface.co/kyujinpy/Sakura-SOLAR-Instruct)
 * [jeonsworld/CarbonVillain-en-10.7B-v1](https://huggingface.co/jeonsworld/CarbonVillain-en-10.7B-v1)
 * [rishiraj/meow](https://huggingface.co/rishiraj/meow)
 * [kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v2](https://huggingface.co/kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v2)

## 🧩 Configuration

```yaml
base_model: kyujinpy/Sakura-SOLAR-Instruct
gate_mode: hidden
experts:
  - source_model: kyujinpy/Sakura-SOLAR-Instruct
    positive_prompts:
    - "chat"
    - "assistant"
    - "tell me"
    - "explain"
    negative_prompts:
    - "mathematics"
    - "reasoning"
  - source_model: jeonsworld/CarbonVillain-en-10.7B-v1
    positive_prompts:
    - "write"
    - "AI"
    - "text"
    - "paragraph"
    negative_prompts:
    - "mathematics"
    - "reasoning"
  - source_model: rishiraj/meow
    positive_prompts:
    - "chat"
    - "say"
    - "what"
    negative_prompts:
    - "mathematics"
    - "reasoning"
  - source_model: kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v2
    positive_prompts:
    - "reason"
    - "math"
    - "mathematics"
    - "solve"
    - "count"
    negative_prompts:
    - "chat"
    - "assistant"
    - "storywriting"
```

## 💻 Usage

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

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/Mixolar-4x7b"

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"])
```
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_shadowml__Mixolar-4x7b)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |74.18|
|AI2 Reasoning Challenge (25-Shot)|71.08|
|HellaSwag (10-Shot)              |88.44|
|MMLU (5-Shot)                    |66.29|
|TruthfulQA (0-shot)              |71.81|
|Winogrande (5-shot)              |83.58|
|GSM8k (5-shot)                   |63.91|