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

Llama 3.2 1B Instruct Abliterated3 by mylesgoose

 »  All LLMs  »  mylesgoose  »  Llama 3.2 1B Instruct Abliterated3   URL Share it on

  10   

Llama 3.2 1B Instruct Abliterated3 is an open-source language model by mylesgoose. Features: 1b LLM, VRAM: 2.5GB, Context: 128K, License: llama3.2, Instruction-Based, LLM Explorer Score: 0.16.

  Arxiv:2204.05149   Conversational   De   En   Endpoints compatible   Es   Facebook   Fr   Hi   Instruct   It   Llama   Llama-3   Meta   Pt   Pytorch   Region:us   Safetensors   Th
Model Card on HF 🤗: https://huggingface.co/mylesgoose/Llama-3.2-1B-Instruct-abliterated3 

Llama 3.2 1B Instruct Abliterated3 Benchmarks

LLME Score: 0.1577
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3.2 1B Instruct Abliterated3 (mylesgoose/Llama-3.2-1B-Instruct-abliterated3)
🌟 Advertise your project 🚀

Llama 3.2 1B Instruct Abliterated3 Parameters and Internals

Model Type 
auto-regressive language model, text generation
Use Cases 
Areas:
commercial, research
Applications:
multilingual dialogue, knowledge retrieval, summarization, mobile AI-powered writing assistants
Primary Use Cases:
instruction-tuned text for assistant-like chat, agentic applications
Limitations:
Prohibited use in violating laws or Acceptable Use Policy, Deployment in languages beyond official support
Considerations:
Developers should tailor safety for specific applications using provided guidelines and tests.
Additional Notes 
Special focus on safety in extreme scenarios like cyber attacks and Bioweapons.
Supported Languages 
English (officially supported), German (officially supported), French (officially supported), Italian (officially supported), Portuguese (officially supported), Hindi (officially supported), Spanish (officially supported), Thai (officially supported)
Training Details 
Data Sources:
Publicly available online data
Data Volume:
Up to 9 trillion tokens
Methodology:
Supervised fine-tuning, Reinforcement learning with human feedback, Knowledge distillation
Context Length:
128000
Training Time:
916k GPU hours
Hardware Used:
Meta's custom built GPU cluster
Model Architecture:
Optimized transformer architecture with Grouped-Query Attention
Safety Evaluation 
Methodologies:
Red-teaming, Safety fine-tuning, Adversarial evaluation datasets
Findings:
Implemented safety mitigations to reduce risks
Risk Categories:
Misinformation, Bias, Cyber Attacks, Child Safety, CBRNE Risks
Ethical Considerations:
Comprehensive safety evaluation conducted to mitigate potential risks.
Responsible Ai Considerations 
Fairness:
Access to diverse language support, promoting inclusivity.
Transparency:
Open community engagement for safety standardization.
Accountability:
Meta retains intellectual property; attribution rights and trademark adherence required.
Mitigation Strategies:
Guidelines for deploying AI with safety guardrails like Llama Guard, Prompt Guard, and Code Shield
Input Output 
Input Format:
Multilingual Text
Accepted Modalities:
text
Output Format:
Multilingual Text and code
Performance Tips:
Deploy with suggested safety guardrails for optimal performance.
Release Notes 
Version:
3.2
Date:
September 25, 2024
Notes:
Many improvements in safety and multilingual capabilities.
LLM NameLlama 3.2 1B Instruct Abliterated3
Repository 🤗https://huggingface.co/mylesgoose/Llama-3.2-1B-Instruct-abliterated3 
Model Size1b
Required VRAM2.5 GB
Updated2026-04-03
Maintainermylesgoose
Model Typellama
Instruction-BasedYes
Model Files  2.5 GB
Supported Languagesen de fr it pt hi es th
Model ArchitectureLlamaForCausalLM
Licensellama3.2
Context Length131072
Model Max Length131072
Transformers Version4.45.0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3.2 1B Instruct Abliterated3

Best Alternatives
Context / RAM
Downloads
Likes
Llama 3.2 1B Instruct128K / 2.5 GB54644911116
Llama 3.2 1B Instruct Bf16128K / 2.5 GB9245
Llama 3.2 1B Instruct128K / 2.5 GB19821791
Llama 3.2 1B Instruct128K / 2.5 GB180
Llama 3.2 OctoThinker INano 1B128K / 3 GB2421
...enchmaxx Llama 3.2 1B Instruct128K / 0 GB134
...p Llama 3.2 1B Instruct Datav2128K / 16.1 GB260
BunderMaxx 1010128K / 2.5 GB141
Llama 3.2 1b OrcaSun V1128K / 3 GB110
... 6 1B Instruct Abliterated LPL128K / 2.5 GB50
Note: green Score (e.g. "73.2") means that the model is better than mylesgoose/Llama-3.2-1B-Instruct-abliterated3.

Rank the Llama 3.2 1B Instruct Abliterated3 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? 52721 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer