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

TowerInstruct 7B V0.2 by Unbabel

 »  All LLMs  »  Unbabel  »  TowerInstruct 7B V0.2   URL Share it on

  2529   
  Arxiv:2402.17733   Autotrain compatible   De   En   Endpoints compatible   Es   Fr   It   Ko   Llama   Nl   Pt   Region:us   Ru   Safetensors   Sharded   Tensorflow   Translation   Zh
Model Card on HF 🤗: https://huggingface.co/Unbabel/TowerInstruct-7B-v0.2 

TowerInstruct 7B V0.2 Benchmarks

LLME Score: 0.20271
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
TowerInstruct 7B V0.2 (Unbabel/TowerInstruct-7B-v0.2)

TowerInstruct 7B V0.2 Parameters and Internals

Model Type 
translation
Use Cases 
Primary Use Cases:
general machine translation, automatic post edition, named-entity recognition, grammatical error correction, paraphrase generation
Limitations:
model may generate problematic outputs like hallucinations, harmful content, false statements
Additional Notes 
The model should not be used as a document-level translator.
Supported Languages 
en (English), de (German), fr (French), zh (Chinese), pt (Portuguese), nl (Dutch), ru (Russian), ko (Korean), it (Italian), es (Spanish)
Training Details 
Data Sources:
TowerBlocks
Methodology:
fine-tuning on the TowerBlocks supervised fine-tuning dataset
Input Output 
Input Format:
ChatML prompt templates
Release Notes 
Version:
TowerInstruct-7B-v0.2
Notes:
Improved document-level translation capabilities
LLM NameTowerInstruct 7B V0.2
Repository 🤗https://huggingface.co/Unbabel/TowerInstruct-7B-v0.2 
Model Size7b
Required VRAM27.1 GB
Updated2025-01-23
MaintainerUnbabel
Model Typellama
Model Files  4.8 GB: 1-of-6   4.9 GB: 2-of-6   4.9 GB: 3-of-6   4.9 GB: 4-of-6   4.9 GB: 5-of-6   2.7 GB: 6-of-6
Supported Languagesen de fr zh pt nl ru ko it es
Model ArchitectureLlamaForCausalLM
Licensecc-by-nc-4.0
Context Length4096
Model Max Length4096
Transformers Version4.37.2
Tokenizer ClassLlamaTokenizer
Padding Token<PAD>
Vocabulary Size32007
Torch Data Typefloat32

Quantized Models of the TowerInstruct 7B V0.2

Model
Likes
Downloads
VRAM
TowerInstruct 7B V0.2 GGUF0412 GB

Best Alternatives to TowerInstruct 7B V0.2

Best Alternatives
Context / RAM
Downloads
Likes
...1M 1000000ctx AEZAKMI 3 1 17021024K / 13.5 GB681
... Qwen2.5llamaify 7B V23.1 200K195K / 15.2 GB37111
LlamaStock 8B128K / 16.1 GB241
SuperNeuralDreadDevil 8B128K / 16.1 GB191
Yarn Llama 2 7B 128K128K / 13.5 GB402739
LLaMA 7B PoSE YaRN 128K128K / 13.5 GB113
LLaMA 7B PoSE Linear 96K96K / 27 GB112
LLaMA 7B PoSE YaRN 96K96K / 13.5 GB101
Chat Llama2 7B 80K80K / 13.8 GB110
Llama2 7B 80K80K / 13.8 GB100
Note: green Score (e.g. "73.2") means that the model is better than Unbabel/TowerInstruct-7B-v0.2.

Rank the TowerInstruct 7B V0.2 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? 41774 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