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

AmberSafe by LLM360

 »  All LLMs  »  LLM360  »  AmberSafe   URL Share it on

  63   
  Arxiv:2312.06550   Autotrain compatible Dataset:pku-alignment/pku-safe...   En   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/LLM360/AmberSafe 

AmberSafe Benchmarks

LLME Score: 0.15269
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
AmberSafe (LLM360/AmberSafe)

AmberSafe Parameters and Internals

Model Type 
Language model
Use Cases 
Areas:
Research, Commercial applications
Applications:
Chatbots, Customer support
Primary Use Cases:
Safe response text generation
Limitations:
Usage restricted to English text
Considerations:
Consult a professional for use in critical applications.
Additional Notes 
Additional support via FastChat available.
Supported Languages 
en (English)
Training Details 
Data Sources:
PKU-Alignment/PKU-SafeRLHF
Methodology:
Supervised fine-tuning and preference learning
Model Architecture:
Same architecture as LLaMA-7B
Safety Evaluation 
Methodologies:
Fine-tuning using a safety-preference dataset
Findings:
AmberSafe showed improved safety performance.
Risk Categories:
Safety, Bias
Ethical Considerations:
Using datasets distinguishing between safe and unsafe responses.
Responsible Ai Considerations 
Fairness:
Dataset filtering ensures safe responses are chosen.
Transparency:
The model and dataset sources are openly available.
Accountability:
LLM360 is accountable for the development process.
Mitigation Strategies:
Using datasets distinguishing between safe and unsafe responses as a strategy to prevent unsafe outputs.
Input Output 
Input Format:
Text prompts
Accepted Modalities:
text
Output Format:
Generated friendly and safe responses
Performance Tips:
Use quantized models for efficiency on less powerful hardware.
Release Notes 
Version:
Q8_0
Notes:
Introduction of 8-bit quantized version for efficiency.
LLM NameAmberSafe
Repository 🤗https://huggingface.co/LLM360/AmberSafe 
Model Size6.7b
Required VRAM13.5 GB
Updated2025-02-22
MaintainerLLM360
Model Typellama
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   3.6 GB: 3-of-3
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to AmberSafe

Best Alternatives
Context / RAM
Downloads
Likes
Gptnoise37128K / 13.5 GB670
Phi 4 RRStock16K / 7.7 GB660
...s Coder6.7b Reflct Adamw Iter216K / 13.5 GB4800
...s Coder6.7b Reflct Adamw Iter116K / 13.5 GB4750
...ir8 Ds Coder6.7b Rmsprop Iter416K / 13.5 GB1830
...s Coder6.7b Reflct Adamw Iter316K / 13.5 GB4300
...ir8 Ds Coder6.7b Rmsprop Iter216K / 13.5 GB1830
Ds Coder6.7b Adamw Iter516K / 13.5 GB4050
...ir8 Ds Coder6.7b Rmsprop Iter316K / 13.5 GB1650
...s Coder6.7b Reflct Adamw Iter416K / 13.5 GB2730
Note: green Score (e.g. "73.2") means that the model is better than LLM360/AmberSafe.

Rank the AmberSafe 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? 43470 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