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

GEITje 7B Ultra by BramVanroy

 »  All LLMs  »  BramVanroy  »  GEITje 7B Ultra   URL Share it on

  1087   
  Arxiv:2412.04092   Alignment-handbook   Autotrain compatible Base model:bramvanroy/geitje-7... Base model:finetune:bramvanroy...   Conversational Dataset:bramvanroy/ultra feedb...   Dpo   Geitje   Mistral   Nl   Region:us   Safetensors   Sharded   Tensorflow   Trl
Model Card on HF 🤗: https://huggingface.co/BramVanroy/GEITje-7B-ultra 

GEITje 7B Ultra Benchmarks

MMLU Pro: 11.24
GPQA: 1.68
MUSR: 1.52
BBH: 12.88
IFEval: 37.23 vs 88 (so35)-57.7%
ARC: 45.48 vs 96.7 (so35)-53%
HellaSwag: 75.5 vs 95.3 (gpt4)-20.8%
MMLU: 50.16 vs 88.3 (so35)-43.2%
TruthfulQA: 53.36 vs 59 (gpt4)-9.6%
WinoGrande: 71.74 vs 87.5 (gpt4)-18%
GSM8K: 19.41 vs 96.4 (so35)-79.9%
MATH Lvl 5: 1.59
LLME Score: 0.23595
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
GEITje 7B Ultra (BramVanroy/GEITje-7B-ultra)

GEITje 7B Ultra Parameters and Internals

Model Type 
text-generation, conversational
Use Cases 
Areas:
research, non-commercial applications
Applications:
chatbot, conversational interface
Primary Use Cases:
Dutch conversational model
Limitations:
Model might generate inaccuracies and potentially offensive content
Supported Languages 
nl (Dutch)
Training Details 
Data Sources:
BramVanroy/ultra_feedback_dutch
Data Volume:
56,137,090 tokens
Methodology:
Based on Mistral 7B, further pretrained on Dutch data
Context Length:
2048
Training Time:
approximately 11 hours
Hardware Used:
Two nodes of four A100 80GB each
Model Architecture:
Based on Mistral 7B and fine-tuned using DPO
Input Output 
Input Format:
Conversations with roles (system, user)
Accepted Modalities:
text
Output Format:
Textual responses
Performance Tips:
Use `load_in_8bit` for reduced memory usage on GPUs; ensure compatibility with flash attention 2 if possible.
LLM NameGEITje 7B Ultra
Repository 🤗https://huggingface.co/BramVanroy/GEITje-7B-ultra 
Base Model(s)  GEITje 7B Ultra Sft   BramVanroy/GEITje-7B-ultra-sft
Model Size7b
Required VRAM14.4 GB
Updated2025-03-22
MaintainerBramVanroy
Model Typemistral
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3   0.0 GB
Supported Languagesnl
Model ArchitectureMistralForCausalLM
Licensecc-by-nc-4.0
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to GEITje 7B Ultra

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB219215
MegaBeam Mistral 7B 512K512K / 14.4 GB388250
SpydazWeb AI HumanAI RP512K / 14.4 GB131
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB377916
MegaBeam Mistral 7B 300K282K / 14.4 GB347116
Hebrew Mistral 7B 200K256K / 30 GB2839915
Astral 256K 7B V2250K / 14.4 GB90
Astral 256K 7B250K / 14.4 GB60
Note: green Score (e.g. "73.2") means that the model is better than BramVanroy/GEITje-7B-ultra.

Rank the GEITje 7B Ultra 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? 45389 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