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

Japanese GPT Neox 3.6B by rinna

 »  All LLMs  »  rinna  »  Japanese GPT Neox 3.6B   URL Share it on

  12161   
  Arxiv:2404.01657   Autotrain compatible   Dataset:cc100   Dataset:mc4   Dataset:wikipedia   Gpt neox   Ja   Lm   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/rinna/japanese-gpt-neox-3.6b 

Japanese GPT Neox 3.6B Benchmarks

ARC: 25 vs 96.7 (so35)-74.1%
HellaSwag: 25.46 vs 95.3 (gpt4)-73.3%
MMLU: 23.12 vs 88.3 (so35)-73.8%
TruthfulQA: 51.45 vs 59 (gpt4)-12.8%
WinoGrande: 50.67 vs 87.5 (gpt4)-42.1%
LLME Score: 0.1747
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Japanese GPT Neox 3.6B (rinna/japanese-gpt-neox-3.6b)
🌟 Advertise your project 🚀

Japanese GPT Neox 3.6B Parameters and Internals

Model Type 
text generation, nlp
Additional Notes 
The model uses sentencepiece tokenizer with specific configurations such as byte fallback and `-add_dummy_prefix` turned off.
Training Details 
Data Sources:
Japanese CC-100, Japanese C4, Japanese Wikipedia
Data Volume:
312.5B tokens
Methodology:
traditional language modelling objective
Model Architecture:
A 36-layer, 2816-hidden-size transformer-based language model
Input Output 
Performance Tips:
set use_fast=False to make certain features function correctly.
LLM NameJapanese GPT Neox 3.6B
Repository 🤗https://huggingface.co/rinna/japanese-gpt-neox-3.6b 
Model Size3.6b
Required VRAM7.4 GB
Updated2025-08-28
Maintainerrinna
Model Typegpt_neox
Model Files  7.4 GB   7.4 GB
Supported Languagesja
Model ArchitectureGPTNeoXForCausalLM
Licensemit
Context Length2048
Model Max Length2048
Tokenizer ClassT5Tokenizer
Padding Token[PAD]
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Japanese GPT Neox 3.6B

Best Alternatives
Context / RAM
Downloads
Likes
...y Jimba Japanese Large Lm 3.6B2K / 7.1 GB50
...rrowSmartPlus 3.6B Instruction2K / 14.3 GB41
...rrowSmartPlus 3.6B Instant Sft2K / 14.3 GB41
...rtPlus 3.6B Instant Sft JHSVer2K / 14.3 GB21
Japanese Large Lm 3.6B2K / 7.2 GB199373
... Large Lm 3.6B Instruction Sft2K / 7.2 GB137125
... GPT Neox 3.6B Instruction Sft2K / 7.4 GB2206105
... GPT Neox 3.6B Instruction Ppo2K / 7.4 GB184473
...T Neox 3.6B Instruction Sft V22K / 7.4 GB152926
...n Sft 4bit 128g Actorder False2K / 2.1 GB62

Rank the Japanese GPT Neox 3.6B 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? 50957 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 v20241124
  Support LLM Explorer