TeenyTinyLlama 160M by nicholasKluge

 ยป  All LLMs  ยป  nicholasKluge  ยป  TeenyTinyLlama 160M   URL Share it on

  Arxiv:2401.16640   Autotrain compatible   Co2 eq emissions Dataset:nicholaskluge/pt-corpu...   Endpoints compatible   Instruct   Jax   Llama   Model-index   Pt   Pytorch   Region:us   Safetensors

TeenyTinyLlama 160M Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
TeenyTinyLlama 160M (nicholasKluge/TeenyTinyLlama-160m)

TeenyTinyLlama 160M Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
research
Primary Use Cases:
Researching language models for low-resource languages
Limitations:
Not suitable for translation or generating text in other languages, Not fine-tuned for downstream contexts
Considerations:
Perform risk and bias assessment before use in real-world applications.
Supported Languages 
Portuguese (Proficient)
Training Details 
Data Sources:
Pt-Corpus Instruct (6.2B tokens)
Data Volume:
6.2B tokens
Methodology:
Transformer-based model pre-trained via causal language modeling
Context Length:
2048
Training Time:
~ 36 hours
Hardware Used:
1 NVIDIA A100-SXM4-40GB
Model Architecture:
Transformer-based
Responsible Ai Considerations 
Fairness:
The model may produce biased or toxic content due to inherited stereotypes from the training data.
Input Output 
Accepted Modalities:
text
Performance Tips:
Ensure proper configuration of repetition penalty and generation parameters like temperature, top-k, and top-p to optimize outputs.
LLM NameTeenyTinyLlama 160M
Repository ๐Ÿค—https://huggingface.co/nicholasKluge/TeenyTinyLlama-160m 
Model Size160m
Required VRAM0.7 GB
Updated2024-12-22
MaintainernicholasKluge
Model Typellama
Instruction-BasedYes
Model Files  0.7 GB   1.3 GB   0.7 GB   0.0 GB
Supported Languagespt
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Padding Token<pad>
Vocabulary Size32000
Torch Data Typebfloat16

Rank the TeenyTinyLlama 160M Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217