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

GPT Neox Japanese 1.4B by stockmark

 »  All LLMs  »  stockmark  »  GPT Neox Japanese 1.4B   URL Share it on

  1520   
  Autotrain compatible   Endpoints compatible   Gpt-neox   Gpt neox   Ja   Japanese   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/stockmark/gpt-neox-japanese-1.4b 

GPT Neox Japanese 1.4B Benchmarks

LLME Score: 0.11697
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
GPT Neox Japanese 1.4B (stockmark/gpt-neox-japanese-1.4b)
🌟 Advertise your project 🚀

GPT Neox Japanese 1.4B Parameters and Internals

Model Type 
text generation
Supported Languages 
Japanese (Native)
Training Details 
Data Sources:
Japanese Web Corpus, Wikipedia (ja), CC100 (ja)
Data Volume:
20B tokens
Methodology:
Trained using HuggingFace Trainer and DeepSpeed (ZeRO-2)
Hardware Used:
8 A100 GPUs (40GB) at ABCI
Model Architecture:
GPT-NeoX
Input Output 
Input Format:
Text prompt
Accepted Modalities:
text
Output Format:
Generated text
Performance Tips:
Use torch.bfloat16 for A100 GPU and torch.float16 for older ones
LLM NameGPT Neox Japanese 1.4B
Repository 🤗https://huggingface.co/stockmark/gpt-neox-japanese-1.4b 
Model Size1.4b
Required VRAM2.9 GB
Updated2025-09-23
Maintainerstockmark
Model Typegpt_neox
Model Files  2.9 GB   2.9 GB
Supported Languagesja
Model ArchitectureGPTNeoXForCausalLM
Licensemit
Context Length1024
Model Max Length1024
Transformers Version4.30.2
Tokenizer ClassGPTNeoXTokenizer
Padding Token<|padding|>
Vocabulary Size50000
Torch Data Typebfloat16

Best Alternatives to GPT Neox Japanese 1.4B

Best Alternatives
Context / RAM
Downloads
Likes
Pythia 1.4B Deduped 8K Base8K / 7.3 GB61
Pythia 1.4B Deduped 4K Base4K / 6.1 GB72
Statichh Pythia 1.4B Sft Bf162K / 5.7 GB7860
Pythia 1.4B2K / 2.9 GB7090325
Pythia 1.4B Sft Full2K / 2.8 GB71
Pythia Delphi Suboptimal2K / 5.7 GB50
Pythia 1.4B Deduped Sharegpt2K / 2.8 GB19122
RCC Ins Reconstruction2K / 6.1 GB81
Pythia 1.4B Deduped Sharegpt2K / 2.8 GB18540
Pythia 1.4B Deduped2K / 2.9 GB949720
Note: green Score (e.g. "73.2") means that the model is better than stockmark/gpt-neox-japanese-1.4b.

Rank the GPT Neox Japanese 1.4B 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? 51536 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