README.md

Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


Draco-8x7B - GGUF
- Model creator: https://huggingface.co/Weyaxi/
- Original model: https://huggingface.co/Weyaxi/Draco-8x7B/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Draco-8x7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q2_K.gguf) | Q2_K | 16.12GB |
| [Draco-8x7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.IQ3_XS.gguf) | IQ3_XS | 18.02GB |
| [Draco-8x7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.IQ3_S.gguf) | IQ3_S | 19.03GB |
| [Draco-8x7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q3_K_S.gguf) | Q3_K_S | 19.03GB |
| [Draco-8x7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.IQ3_M.gguf) | IQ3_M | 19.96GB |
| [Draco-8x7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q3_K.gguf) | Q3_K | 21.0GB |
| [Draco-8x7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q3_K_M.gguf) | Q3_K_M | 21.0GB |
| [Draco-8x7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q3_K_L.gguf) | Q3_K_L | 22.51GB |
| [Draco-8x7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.IQ4_XS.gguf) | IQ4_XS | 23.63GB |
| [Draco-8x7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q4_0.gguf) | Q4_0 | 24.63GB |
| [Draco-8x7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.IQ4_NL.gguf) | IQ4_NL | 24.91GB |
| [Draco-8x7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q4_K_S.gguf) | Q4_K_S | 24.91GB |
| [Draco-8x7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q4_K.gguf) | Q4_K | 26.49GB |
| [Draco-8x7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q4_K_M.gguf) | Q4_K_M | 26.49GB |
| [Draco-8x7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q4_1.gguf) | Q4_1 | 27.32GB |
| [Draco-8x7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q5_0.gguf) | Q5_0 | 30.02GB |
| [Draco-8x7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q5_K_S.gguf) | Q5_K_S | 30.02GB |
| [Draco-8x7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q5_K.gguf) | Q5_K | 30.95GB |
| [Draco-8x7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q5_K_M.gguf) | Q5_K_M | 30.95GB |
| [Draco-8x7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q5_1.gguf) | Q5_1 | 32.71GB |
| [Draco-8x7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/blob/main/Draco-8x7B.Q6_K.gguf) | Q6_K | 35.74GB |
| [Draco-8x7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/Weyaxi_-_Draco-8x7B-gguf/tree/main/) | Q8_0 | 46.22GB |




Original model description:
---
license: apache-2.0
tags:
- moe
- openchat
- hermes
- dolphin
- bagel
model-index:
- name: Draco-8x7B
  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: 65.02
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PulsarAI/Draco-8x7B
      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: 85.24
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PulsarAI/Draco-8x7B
      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: 64.96
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PulsarAI/Draco-8x7B
      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: 62.65
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PulsarAI/Draco-8x7B
      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: 80.66
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PulsarAI/Draco-8x7B
      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: 66.79
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PulsarAI/Draco-8x7B
      name: Open LLM Leaderboard
---
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6468ce47e134d050a58aa89c/VWIJplnya5L7wmGxK4Lut.jpeg)

# 💫 Draco-8x7B

This is the model for Draco-8x7B. I used [this repo](https://bit.ly/weyaxi-moe-repo) to make this MOE model.

This model's experts are not using any merged models.

# 📚 Other branches (Number of Experts Per Token)

Other branches that this repository contains differ only slightly (from a git diff perspective) in terms of the number of experts per token.

Usually, a higher value for the number of experts per token will result in better performance, but it may also lead to increased inference time.

| Number of experts per token  | Link of the branch                                                                         |
| ---------------------------- | -------------------------------------------------------------------------------------------|
| 2                            | [Main](https://huggingface.co/Weyaxi/Draco-8x7B/tree/main)                                 |
| 3                            | [3-experts-per-token](https://huggingface.co/Weyaxi/Draco-8x7B/tree/3-experts-per-token)   |
| 4                            | [4-experts-per-token](https://huggingface.co/Weyaxi/Draco-8x7B/tree/4-experts-per-token)   |
| 6                            | [6-experts-per-token](https://huggingface.co/Weyaxi/Draco-8x7B/tree/6-experts-per-token)   |
| 8                            | [8-experts-per-token](https://huggingface.co/Weyaxi/Draco-8x7B/tree/8-experts-per-token)   |


# 💬 Prompt Template(s):

This model includes many models, so providing only one prompt template is not enough. You can use and try these prompt templates and decide which works best for you.

**Note:** The current chat template in the tokenizer config is set to [openchat-3.5-0106](https://huggingface.co/openchat/openchat-3.5-0106)'s chat template.

**Note 2:** It is also important to note that [jondurbin/bagel-dpo-7b-v0.1](https://huggingface.co/jondurbin/bagel-dpo-7b-v0.1) is using many prompt templates other than I provided. You can visit [jondurbin/bagel-dpo-7b-v0.1](https://huggingface.co/jondurbin/bagel-dpo-7b-v0.1) to learn more about this templates.

### GPT4 Correct

Used in [openchat/openchat-3.5-0106](https://huggingface.co/openchat/openchat-3.5-0106), [beowolx/CodeNinja-1.0-OpenChat-7B](https://huggingface.co/beowolx/CodeNinja-1.0-OpenChat-7B)

```
GPT4 Correct User: {user}<|end_of_turn|>GPT4 Correct Assistant: {asistant}<|end_of_turn|>
```

### ChatML:

Used in [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B), [jondurbin/bagel-dpo-7b-v0.1](https://huggingface.co/jondurbin/bagel-dpo-7b-v0.1), [cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser](https://huggingface.co/cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser), [senseable/WestLake-7B-v2](https://huggingface.co/senseable/WestLake-7B-v2)

```
<|im_start|>system
{system}<|im_end|>
<|im_start|>user
{user}<|im_end|>
<|im_start|>assistant
{asistant}<|im_end|>
```

### Math Alpaca

Used in [meta-math/MetaMath-Mistral-7B](https://huggingface.co/meta-math/MetaMath-Mistral-7B)

```
Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{instruction}

### Response: Let's think step by step.
```

# 🛠️ Yaml Config

<details><summary>See config</summary>

```yaml
base_model: openchat/openchat-3.5-0106
gate_mode: hidden
dtype: bfloat16

experts:
  - source_model: openchat/openchat-3.5-0106
    positive_prompts: # General (Mistral finetune)
    - "chat"
    - "assistant"
    - "tell me"
    - "explain"

  - source_model: teknium/OpenHermes-2.5-Mistral-7B
    positive_prompts: # General (Mistral finetune)
    - "interact"
    - "converse"
    - "respond"
    - "express"

  - source_model: jondurbin/bagel-dpo-7b-v0.1
    positive_prompts: # Science (Mistral finetune)
    - "science"
    - "biology"
    - "chemistry"
    - "physics"
    - "Newton's laws"
    - "scientific method"
    - "periodic table"
    - "photosynthesis process"

  - source_model: meta-math/MetaMath-Mistral-7B
    positive_prompts: # Math (Mistral finetune)
    - "reason"
    - "math"
    - "mathematics"
    - "solve"
    - "count"

  - source_model: cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser
    positive_prompts: # Uncensored (Mistral finetune)
    - "dolphin"
    - "uncensored"
    - "unbiased"
    - "unfiltered"
    - "unrestricted"
    - "offensive"

  - source_model: beowolx/CodeNinja-1.0-OpenChat-7B
    positive_prompts: # Code (openchat-3.5-1210 finetune)
    - "code"
    - "script"
    - "python"
    - "javascript"
    - "programming"
    - "algorithm"

  - source_model: senseable/WestLake-7B-v2
    positive_prompts: # Roleplay (Unknown finetune)
    - "storywriting"
    - "write"
    - "scene"
    - "story"
    - "character"
    - "act as"
    - "you are"

  - source_model: snorkelai/Snorkel-Mistral-PairRM-DPO
    positive_prompts: # Question Answering (? Mistral-7B-Instruct-v0.2 finetune ?)
    - "what happens"
    - "what is"
    - "what can"
    - "why"
    - "who"
    - "can a"
```

</details><br>

# 🔄 Quantizationed versions

Quantizationed versions of this model is available thanks to [TheBloke](https://hf.co/TheBloke).

##### GPTQ

- [TheBloke/Draco-8x7B-GPTQ](https://huggingface.co/TheBloke/Draco-8x7B-GPTQ)

##### GGUF

- [TheBloke/Draco-8x7B-GGUF](https://huggingface.co/TheBloke/Draco-8x7B-GGUF)

##### AWQ

- [TheBloke/Draco-8x7B-AWQ](https://huggingface.co/TheBloke/Draco-8x7B-AWQ)

If you would like to support me:

[☕ Buy Me a Coffee](https://www.buymeacoffee.com/weyaxi)
# [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_PulsarAI__Draco-8x7B)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |70.89|
|AI2 Reasoning Challenge (25-Shot)|65.02|
|HellaSwag (10-Shot)              |85.24|
|MMLU (5-Shot)                    |64.96|
|TruthfulQA (0-shot)              |62.65|
|Winogrande (5-shot)              |80.66|
|GSM8k (5-shot)                   |66.79|