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

Llama 3.1 8B Instruct Ortho V7 by lodrick-the-lafted

 »  All LLMs  »  lodrick-the-lafted  »  Llama 3.1 8B Instruct Ortho V7   URL Share it on

  9   
  Autotrain compatible   Conversational   Endpoints compatible   Instruct   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/lodrick-the-lafted/llama-3.1-8b-instruct-ortho-v7 

Llama 3.1 8B Instruct Ortho V7 Benchmarks

MMLU Pro: 10.82
GPQA: 3.02
MUSR: 4.73
BBH: 14.44
IFEval: 35.15 vs 88 (so35)-60.1%
MATH Lvl 5: 2.72
LLME Score: 0.22013
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3.1 8B Instruct Ortho V7 (lodrick-the-lafted/llama-3.1-8b-instruct-ortho-v7)

Llama 3.1 8B Instruct Ortho V7 Parameters and Internals

Model Type 
text generation
Additional Notes 
Each attempt uses different vectors and varies in refusal boundaries. The method requires filtering for interesting control vectors.
Release Notes 
Version:
v3
Notes:
Attempt at orthogonalization of llama-3.1-8b-instruct using a method from 'Mechanistically Eliciting Latent Behaviors in Language Models'.
Version:
v4
Notes:
Continued attempts with different vectors. Minor variations in refusal boundaries.
Version:
v5
Notes:
Further variations attempting orthogonalization.
Version:
v6
Notes:
Additional orthogonalization attempts.
Version:
v7
Notes:
Latest in the series of orthogonalization adjustments with continued issues in jailbreaking.
LLM NameLlama 3.1 8B Instruct Ortho V7
Repository 🤗https://huggingface.co/lodrick-the-lafted/llama-3.1-8b-instruct-ortho-v7 
Model Size8b
Required VRAM16.1 GB
Updated2025-05-15
Maintainerlodrick-the-lafted
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
Licensewtfpl
Context Length131072
Model Max Length131072
Transformers Version4.44.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|finetune_right_pad_id|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3.1 8B Instruct Ortho V7

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB5687104
UltraLong Thinking4192K / 16.1 GB2712
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB102715
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB499441
Zero Llama 3.1 8B Beta61048K / 16.1 GB1051
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
...dger Nu Llama 3.1 8B UltraLong1048K / 16.2 GB532
....1 1million Ctx Dark Planet 8B1048K / 32.3 GB392
Note: green Score (e.g. "73.2") means that the model is better than lodrick-the-lafted/llama-3.1-8b-instruct-ortho-v7.

Rank the Llama 3.1 8B Instruct Ortho V7 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? 47368 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