LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Llama 3 8B Instruct Abliterated V2 by cognitivecomputations

 »  All LLMs  »  cognitivecomputations  »  Llama 3 8B Instruct Abliterated V2   URL Share it on

  5134   
  Autotrain compatible   Conversational   Endpoints compatible   Instruct   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2 

Llama 3 8B Instruct Abliterated V2 Benchmarks

ARC: 59.73 vs 96.7 (so35)-38.2%
HellaSwag: 79.29 vs 95.3 (gpt4)-16.8%
MMLU: 67.43 vs 88.3 (so35)-23.6%
TruthfulQA: 43.97 vs 59 (gpt4)-25.5%
WinoGrande: 74.27 vs 87.5 (gpt4)-15.1%
GSM8K: 71.34 vs 96.4 (so35)-26%
LLME Score: 0.23169
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3 8B Instruct Abliterated V2 (cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2)

Llama 3 8B Instruct Abliterated V2 Parameters and Internals

Model Type 
language model
Additional Notes 
This model has certain weights manipulated to inhibit the model's ability to express refusal. It is notably better at directly answering requests without producing disclaimers. It may still refuse to answer some requests, as the methodology is new and untested.
Training Details 
Methodology:
The methodology used to generate this model involves an orthogonalization process to manipulate certain weights to inhibit the model's ability to express refusal.
Model Architecture:
Llama-3
LLM NameLlama 3 8B Instruct Abliterated V2
Repository 🤗https://huggingface.co/cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2 
Model Size8b
Required VRAM16.1 GB
Updated2025-04-19
Maintainercognitivecomputations
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.41.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3 8B Instruct Abliterated V2

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB63871
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB115614
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB171836
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
...a 3 8B Instruct Gradient 1048K1024K / 16.1 GB25079679
A4.21024K / 16.1 GB1350
A3.21024K / 16.1 GB1340
A2.21024K / 16.1 GB1340
Note: green Score (e.g. "73.2") means that the model is better than cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2.

Rank the Llama 3 8B Instruct Abliterated V2 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 46445 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer