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 Open Ko 8B Gguf by teddylee777

 »  All LLMs  »  teddylee777  »  Llama 3 Open Ko 8B Gguf   URL Share it on

  2522   
  Arxiv:2310.04799   Autotrain compatible   Conversational   En   Endpoints compatible   Facebook   Gguf   Ko   Llama   Llama-3   Llama-3-ko   Meta   Pytorch   Q4   Quantized   Region:us
Model Card on HF 🤗: https://huggingface.co/teddylee777/Llama-3-Open-Ko-8B-gguf 

Llama 3 Open Ko 8B Gguf Benchmarks

ARC: 55.72 vs 96.7 (so35)-42.4%
HellaSwag: 74.92 vs 95.3 (gpt4)-21.4%
MMLU: 53.78 vs 88.3 (so35)-39.1%
TruthfulQA: 46.03 vs 59 (gpt4)-22%
WinoGrande: 72.53 vs 87.5 (gpt4)-17.1%
GSM8K: 22.29 vs 96.4 (so35)-76.9%
LLME Score: 0.26225
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3 Open Ko 8B Gguf (teddylee777/Llama-3-Open-Ko-8B-gguf)

Llama 3 Open Ko 8B Gguf Parameters and Internals

Model Type 
text generation, dialogue optimization
Use Cases 
Areas:
Commercial and Research Applications
Primary Use Cases:
Assistant-like chat, Various natural language generation tasks
Limitations:
Any use violating applicable laws or the Acceptable Use Policy is prohibited.
Considerations:
Thorough safety testing and tuning recommended before deployment.
Supported Languages 
Korean (High proficiency), English (High proficiency)
Training Details 
Data Sources:
Open-Solar-Ko Dataset
Data Volume:
Over 60GB of deduplicated texts
Methodology:
New Llama-3 tokenizer
Context Length:
4096
Hardware Used:
TPUv5e-256
Model Architecture:
Optimized transformer architecture
Safety Evaluation 
Ethical Considerations:
Emphasizes openness, inclusivity, and helpfulness.
Responsible Ai Considerations 
Fairness:
Aimed for inclusivity and helpfulness across diverse backgrounds.
Mitigation Strategies:
Developers encouraged to use Meta Llama Guard 2 and Code Shield.
Input Output 
Input Format:
Text input
Accepted Modalities:
text
Output Format:
Text and code generation
Release Notes 
Version:
Llama-3-Open-Ko-8B
Date:
2024.04.24
Notes:
Release of Llama-3-Open-Ko-8B model.
LLM NameLlama 3 Open Ko 8B Gguf
Repository 🤗https://huggingface.co/teddylee777/Llama-3-Open-Ko-8B-gguf 
Base Model(s)  Llama 3 Open Ko 8B   beomi/Llama-3-Open-Ko-8B
Model Size8b
Required VRAM4.7 GB
Updated2024-12-22
Maintainerteddylee777
Model Typellama
Model Files  16.1 GB   4.7 GB   5.1 GB   4.9 GB   4.7 GB   5.6 GB   5.7 GB   5.6 GB   6.6 GB   8.5 GB
Supported Languagesen ko
GGUF QuantizationYes
Quantization Typegguf|q4|q4_k|q5_k
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.1
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3 Open Ko 8B Gguf

Best Alternatives
Context / RAM
Downloads
Likes
...truct Gradient 1048K IMat GGUF1024K / 2 GB2886
...B Instruct Gradient 1048K GGUF1024K / 3.2 GB1743
Unhinged Llama3 8B 524K512K / 26.5 GB250
Llama 3 8B Instruct 262K GGUF256K / 3.2 GB1682
... 8B Instruct Reasoner 1o1 V0.3128K / 16.1 GB2247
Reflection Llama 3.1 8B128K / 16.1 GB305117
Alpaca Llama3.1 8B128K / 16.1 GB60
...lama 3.1 Cantonese 8B Instruct128K / 16.1 GB6385
... Horizon AI Korean Advanced 8B128K / 16.1 GB7920
ProductLlama V2128K / 16.1 GB210
Note: green Score (e.g. "73.2") means that the model is better than teddylee777/Llama-3-Open-Ko-8B-gguf.

Rank the Llama 3 Open Ko 8B Gguf 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? 40066 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 v20241217
  Support LLM Explorer