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

EastAsia 4x7B MoE Experiment by Heng666

 »  All LLMs  »  Heng666  »  EastAsia 4x7B MoE Experiment   URL Share it on

  1746   
  Augmxnt/shisa-7b-v1   Autotrain compatible   Beomi/open-solar-ko-10.7b   Endpoints compatible   Instruct   Ja   Ko   Lazymergekit Mediatek-research/breeze-7b-in...   Merge   Mergekit   Mixtral   Model-index   Moe   Region:us   Safetensors   Sharded   Tensorflow   Tw   Zh
Model Card on HF 🤗: https://huggingface.co/Heng666/EastAsia-4x7B-Moe-experiment 

EastAsia 4x7B MoE Experiment Benchmarks

ARC: 39.51 vs 96.7 (so35)-59.1%
HellaSwag: 48.92 vs 95.3 (gpt4)-48.7%
MMLU: 56.2 vs 88.3 (so35)-36.4%
TruthfulQA: 49.83 vs 59 (gpt4)-15.5%
WinoGrande: 58.09 vs 87.5 (gpt4)-33.6%
GSM8K: 0.15 vs 96.4 (so35)-99.8%
LLME Score: 0.21309
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
EastAsia 4x7B MoE Experiment (Heng666/EastAsia-4x7B-Moe-experiment)

EastAsia 4x7B MoE Experiment Parameters and Internals

Model Type 
Mixture of Experts, text generation
Additional Notes 
EastAsia-4x7B-Moe-experiment is a Mixture of Experts model created using LazyMergekit and combines several base models.
Supported Languages 
zh (unknown proficiency), ja (unknown proficiency), ko (unknown proficiency), tw (unknown proficiency)
LLM NameEastAsia 4x7B MoE Experiment
Repository 🤗https://huggingface.co/Heng666/EastAsia-4x7B-Moe-experiment 
Model Size18.5b
Required VRAM37.1 GB
Updated2025-03-14
MaintainerHeng666
Model Typemixtral
Instruction-BasedYes
Model Files  9.9 GB: 1-of-4   10.0 GB: 2-of-4   10.0 GB: 3-of-4   7.2 GB: 4-of-4
Supported Languageszh ja ko tw
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to EastAsia 4x7B MoE Experiment

Best Alternatives
Context / RAM
Downloads
Likes
Rava 3x7B V0.132K / 37.1 GB171

Rank the EastAsia 4x7B MoE Experiment 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? 45019 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