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

ELYZA Japanese Llama 2 13B Instruct by elyza

 »  All LLMs  »  elyza  »  ELYZA Japanese Llama 2 13B Instruct   URL Share it on

  2316   
  Arxiv:2307.09288   Autotrain compatible   En   Endpoints compatible   Instruct   Ja   Llama   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-instruct 

ELYZA Japanese Llama 2 13B Instruct Benchmarks

ARC: 58.36 vs 96.7 (so35)-39.6%
HellaSwag: 82.2 vs 95.3 (gpt4)-13.7%
MMLU: 55.65 vs 88.3 (so35)-37%
TruthfulQA: 42.4 vs 59 (gpt4)-28.1%
WinoGrande: 75.22 vs 87.5 (gpt4)-14%
GSM8K: 14.48 vs 96.4 (so35)-85%
LLME Score: 0.17938
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
ELYZA Japanese Llama 2 13B Instruct (elyza/ELYZA-japanese-Llama-2-13b-instruct)
🌟 Advertise your project 🚀

ELYZA Japanese Llama 2 13B Instruct Parameters and Internals

Model Type 
text generation
Additional Notes 
The model is meant to enhance Japanese language capabilities based on the Llama 2 model, tailored for specific assistance tasks.
Supported Languages 
ja (Advanced), en (Advanced)
Training Details 
Data Sources:
Llama 2 base data
Methodology:
Additional pre-training for enhanced Japanese capability
Input Output 
Input Format:
[INST] {system} [/] {prompts} [/]
Accepted Modalities:
text
Output Format:
Textual response
Performance Tips:
Use device_map='auto' and ensure use of optimized configurations for best performance.
LLM NameELYZA Japanese Llama 2 13B Instruct
Repository 🤗https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b-instruct 
Model Size13b
Required VRAM26 GB
Updated2025-08-18
Maintainerelyza
Model Typellama
Instruction-BasedYes
Model Files  9.9 GB: 1-of-3   9.9 GB: 2-of-3   6.2 GB: 3-of-3
Supported Languagesja en
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.34.1
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to ELYZA Japanese Llama 2 13B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
NexusRaven V2 13B16K / 26 GB4303469
CodeLlama 13B Instruct Hf16K / 26 GB12751154
CodeLlama 13B MORepair16K / 26 GB122
CodeLlama 13B Instruct Hf16K / 26 GB164225
TableLLM 13B16K / 26 GB101029
NexusRaven 13B16K / 26 GB36104
Panda Coder 13B16K / 26 GB1713
... Llama 2 13B Instruct Text2sql16K / 26 GB3527
Gen Sim16K / 0.3 GB72
Llama 3 13B Instruct Ft8K / 26.1 GB52
Note: green Score (e.g. "73.2") means that the model is better than elyza/ELYZA-japanese-Llama-2-13b-instruct.

Rank the ELYZA Japanese Llama 2 13B Instruct 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? 50729 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