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

Llama 3 Aplite Instruct 4x8B MoE by raincandy-u

 »  All LLMs  »  raincandy-u  »  Llama 3 Aplite Instruct 4x8B MoE   URL Share it on

  462   
  Autotrain compatible Base model:finetune:meta-llama... Base model:meta-llama/meta-lla...   Code   Conversational   En   Endpoints compatible   Facebook   Instruct   Llama   Llama-3   Meta   Mixtral   Model-index   Moe   Pytorch   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/raincandy-u/Llama-3-Aplite-Instruct-4x8B-MoE 

Llama 3 Aplite Instruct 4x8B MoE Benchmarks

ARC: 61.86 vs 96.7 (so35)-36%
HellaSwag: 81.08 vs 95.3 (gpt4)-14.9%
MMLU: 67.47 vs 88.3 (so35)-23.6%
TruthfulQA: 51.87 vs 59 (gpt4)-12.1%
WinoGrande: 77.82 vs 87.5 (gpt4)-11.1%
GSM8K: 61.87 vs 96.4 (so35)-35.8%
LLME Score: 0.22776
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3 Aplite Instruct 4x8B MoE (raincandy-u/Llama-3-Aplite-Instruct-4x8B-MoE)

Llama 3 Aplite Instruct 4x8B MoE Parameters and Internals

Model Type 
text-generation
Additional Notes 
This model is a research experiment and may generate incorrect or harmful content. Outputs should not be taken as factual or representative of the views of the model's creator or any individual.
Training Details 
Data Sources:
Meta-Llama-3-8B-Instruct, Llama3-8B-OpenHermes-DPO, Llama-3-SLERP-8B, hf-llama3-8b-orpo-v0.0
LLM NameLlama 3 Aplite Instruct 4x8B MoE
Repository 🤗https://huggingface.co/raincandy-u/Llama-3-Aplite-Instruct-4x8B-MoE 
Base Model(s)  Meta Llama 3 8B Instruct   meta-llama/Meta-Llama-3-8B-Instruct
Model Size24.9b
Required VRAM50 GB
Updated2024-12-22
Maintainerraincandy-u
Model Typemixtral
Instruction-BasedYes
Model Files  9.9 GB: 1-of-6   10.0 GB: 2-of-6   10.0 GB: 3-of-6   9.9 GB: 4-of-6   9.1 GB: 5-of-6   1.1 GB: 6-of-6
Supported Languagesen
Model ArchitectureMixtralForCausalLM
Licenseother
Context Length8192
Model Max Length8192
Transformers Version4.39.3
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|begin_of_text|>
Vocabulary Size128256
Torch Data Typefloat16

Best Alternatives to Llama 3 Aplite Instruct 4x8B MoE

Best Alternatives
Context / RAM
Downloads
Likes
L3.1 MoE 4x8B V0.2128K / 50.1 GB352
OpenCrystal V4 L3 4x8B8K / 50 GB92
L3 Arcania 4x8b8K / 50 GB40222
Llama 3 8B Instruct MoE 28K / 50 GB110
Llama 3 4x8B8K / 49.9 GB130
Note: green Score (e.g. "73.2") means that the model is better than raincandy-u/Llama-3-Aplite-Instruct-4x8B-MoE.

Rank the Llama 3 Aplite Instruct 4x8B MoE 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? 40066 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 v20241217
  Support LLM Explorer