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

SmolLM 135M De by LemiSt

 »  All LLMs  »  LemiSt  »  SmolLM 135M De   URL Share it on

  1273   
Base model:finetune:huggingfac... Base model:huggingfacetb/smoll...   Dataset:almanach/halvest Dataset:d4ve-r/terra-xplain-cc... Dataset:devngho/culturax-mini-...   Dataset:djstrong/oscar-small   Dataset:lemist/gutenberg de Dataset:maxidl/finenews-unfilt...   Dataset:wikimedia/wikipedia   De   Endpoints compatible   Feature-extraction   Llama   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/LemiSt/SmolLM-135M-de 

SmolLM 135M De Benchmarks

LLME Score: 0.25003
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
SmolLM 135M De (LemiSt/SmolLM-135M-de)

SmolLM 135M De Parameters and Internals

Model Type 
Large Language Model (Llama architecture)
Use Cases 
Areas:
small experimentation
Applications:
benchmarking datasets
Limitations:
Will output blatantly wrong information, Possible generation of inappropriate content, Not recommended for production use
Considerations:
Consider further fine tuning and preference optimization before use
Additional Notes 
The model is primarily for experimentation and benchmarking rather than production use. Outputs correct German primarily.
Supported Languages 
German (advanced)
Training Details 
Data Sources:
devngho/culturax-mini-nonshuffled, maxidl/FineNews-unfiltered, djstrong/oscar-small, LemiSt/gutenberg_de, almanach/HALvest, wikimedia/wikipedia, D4ve-R/terra-xplain-cc-de
Data Volume:
about 6 billion German-language tokens
Methodology:
trained with axolotl, using full fine tuning
Context Length:
2048
Training Time:
nearly 2 epochs
Model Architecture:
Llama architecture
LLM NameSmolLM 135M De
Repository 🤗https://huggingface.co/LemiSt/SmolLM-135M-de 
Base Model(s)  SmolLM 135M   HuggingFaceTB/SmolLM-135M
Model Size135m
Required VRAM0.5 GB
Updated2025-02-22
MaintainerLemiSt
Model Typellama
Model Files  0.5 GB
Supported Languagesde
Model ArchitectureLlamaModel
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.44.2
Tokenizer ClassGPT2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size49152
Torch Data Typefloat32
Errorsreplace

Rank the SmolLM 135M De 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