File size: 5,458 Bytes
e88684d 732f7bb e88684d 732f7bb e88684d 732f7bb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 |
---
language:
- en
license: cc-by-nc-4.0
tags:
- mixtral
- uncensored
- high-intelligence
model-index:
- name: MixtralOrochi8x7B
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: 70.31
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=smelborp/MixtralOrochi8x7B
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: 86.1
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=smelborp/MixtralOrochi8x7B
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: 70.13
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=smelborp/MixtralOrochi8x7B
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: 63.99
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=smelborp/MixtralOrochi8x7B
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: 79.87
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=smelborp/MixtralOrochi8x7B
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: 17.29
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=smelborp/MixtralOrochi8x7B
name: Open LLM Leaderboard
---
# Orochi
<img src="https://huggingface.co/smelborp/MixtralOrochi8x7B/resolve/main/orochi.png" width="600" />
## Overview
Orochi is a cutting-edge language model based on the Mixtral architecture developed by Mistral. It represents a sophisticated merge of several prominent models, including Mixtral instruct, Noromaid, OpenBuddy, and several others, using mergekit with the DARE merge method. This model aims to provide highly intelligent responses unrestricted by content limitations. The name "Orochi" references the mythical Yamata-no-Orochi, symbolizing the model's multifaceted and powerful capabilities.
## Goals
- **Uncensored Content**: To provide unrestricted and comprehensive responses across various domains.
- **High Intelligence**: Leverage the combined knowledge and capabilities of the merged models to deliver insightful and accurate information.
- **Innovation in Language Modeling**: Push the boundaries of what's possible in natural language understanding and generation.
## Model Details
- **Architecture**: Mixtral, a Mixture of Experts model, underlies Orochi's design, enabling it to specialize and optimize its responses across different tasks and topics.
- **Merge Strategy**: Utilizing mergekit and the DARE method, Orochi integrates aspects of various models to enhance its performance and capabilities.
## Usage
Due to its uncensored nature, Orochi is best utilized in environments where intelligent, unrestricted dialogue is necessary. Users are encouraged to implement their own content moderation or alignment strategies appropriate for their use case.
## Ethical Considerations
As an uncensored model, Orochi may generate content that is unsuitable for all audiences. Users are advised to consider the implications of using such a model and to implement suitable safeguards and ethical guidelines.
## Acknowledgements
Orochi is a product of numerous contributions from the fields of machine learning and language modeling. Special thanks to the teams behind Mixtral, mergekit, and all the individual models integrated into Orochi.
---
# [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_smelborp__MixtralOrochi8x7B)
| Metric |Value|
|---------------------------------|----:|
|Avg. |64.62|
|AI2 Reasoning Challenge (25-Shot)|70.31|
|HellaSwag (10-Shot) |86.10|
|MMLU (5-Shot) |70.13|
|TruthfulQA (0-shot) |63.99|
|Winogrande (5-shot) |79.87|
|GSM8k (5-shot) |17.29|
|