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

LIMA 13B by heegyu

 »  All LLMs  »  heegyu  »  LIMA 13B   URL Share it on

  1247   
  Autotrain compatible   Endpoints compatible   Llama   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/heegyu/LIMA-13b 

LIMA 13B Benchmarks

LLME Score: 0.13647
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
LIMA 13B (heegyu/LIMA-13b)

LIMA 13B Parameters and Internals

Model Type 
auto-regressive, transformer architecture
Use Cases 
Primary Use Cases:
Research on large language models, Exploring applications such as question answering and reading comprehension, Evaluating and mitigating biases, Determining capabilities and limitations of models
Limitations:
Base model not suitable for downstream applications without risk evaluation
Supported Languages 
en (High proficiency), others (Included 20 languages, mainly supports English)
Training Details 
Data Sources:
CCNet, C4, GitHub, Wikipedia, Books, ArXiv, Stack Exchange
Data Volume:
1T tokens with different breakdowns for different model sizes
Model Architecture:
transformer architecture
Responsible Ai Considerations 
Fairness:
Model reflects biases from web sources. Evaluated biases include gender, religion, race, sexual orientation, age, nationality, disability, physical appearance, and socioeconomic status.
Transparency:
Model trained using web-sourced data which may contain biased and harmful content.
Accountability:
Use GitHub repository to raise questions or comments.
Mitigation Strategies:
Filtered data based on proximity to Wikipedia text using a Kneser-Ney language model and fastText linear classifier.
LLM NameLIMA 13B
Repository 🤗https://huggingface.co/heegyu/LIMA-13b 
Model Size13b
Required VRAM42 GB
Updated2025-02-05
Maintainerheegyu
Model Typellama
Model Files  1.0 GB: 0-of-41   1.0 GB: 1-of-41   1.0 GB: 2-of-41   1.0 GB: 3-of-41   1.0 GB: 4-of-41   1.0 GB: 5-of-41   1.0 GB: 6-of-41   1.0 GB: 7-of-41   1.0 GB: 8-of-41   1.0 GB: 9-of-41   1.0 GB: 10-of-41   1.0 GB: 11-of-41   1.0 GB: 12-of-41   1.0 GB: 13-of-41   1.0 GB: 14-of-41   1.0 GB: 15-of-41   1.0 GB: 16-of-41   1.0 GB: 17-of-41   1.0 GB: 18-of-41   1.0 GB: 19-of-41   1.0 GB: 20-of-41   1.0 GB: 21-of-41   1.0 GB: 22-of-41   1.0 GB: 23-of-41   1.0 GB: 24-of-41   1.0 GB: 25-of-41   1.0 GB: 26-of-41   1.0 GB: 27-of-41   1.0 GB: 28-of-41   1.0 GB: 29-of-41   1.0 GB: 30-of-41   1.0 GB: 31-of-41   1.0 GB: 32-of-41   1.0 GB: 33-of-41   1.0 GB: 34-of-41   1.0 GB: 35-of-41   1.0 GB: 36-of-41   1.0 GB: 37-of-41   1.0 GB: 38-of-41   1.0 GB: 39-of-41   1.0 GB: 40-of-41   1.0 GB: 41-of-41
Model ArchitectureLLaMAForCausalLM
Licenseother
Transformers Version4.27.0.dev0
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to LIMA 13B

Best Alternatives
Context / RAM
Downloads
Likes
Llm Jp 13B V2.04K / 27.4 GB68815
Decapoda Research Llama 13B0K / 41 GB50
Alpaca 13B0K / 52.1 GB2347108
Llama 13B0K / 42 GB151
Vicuna20K / 0 GB616
Llama 13B0K / 42 GB93
... X Alpaca 13B Native 4bit 128g0K / 7.9 GB1498735
... X Alpaca 13B Native 4bit 128g0K / 8.1 GB42
Llama 13B 4bit Hf0K / 7 GB62
Llama 13B 4bit Gr1280K / 7.5 GB52
Note: green Score (e.g. "73.2") means that the model is better than heegyu/LIMA-13b.

Rank the LIMA 13B 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? 42577 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