|
gptkbp:instanceOf
|
large language model
|
|
gptkbp:architecture
|
gptkb:transformation
|
|
gptkbp:availableOn
|
gptkb:Hugging_Face
gptkb:AWS
gptkb:Google_Cloud
gptkb:Meta_AI_website
gptkb:Azure
|
|
gptkbp:bench
|
gptkb:BBH
gptkb:GSM8K
gptkb:HellaSwag
gptkb:HumanEval
gptkb:MMLU
gptkb:OpenBookQA
gptkb:PIQA
gptkb:TruthfulQA
gptkb:ARC
Winogrande
|
|
gptkbp:category
|
gptkb:model
open-weight model
chatbot model
|
|
gptkbp:context
|
8,192 tokens
|
|
gptkbp:developedBy
|
gptkb:Meta_AI
|
|
gptkbp:fineTunedWith
|
gptkb:public_speaker
chat applications
instruction following
|
|
gptkbp:hasModel
|
decoder-only transformer
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Llama 3 8B Chat
|
|
gptkbp:intendedUse
|
chatbot
conversational AI
|
|
gptkbp:language
|
English
|
|
gptkbp:length
|
8,192 tokens
|
|
gptkbp:license
|
Meta Llama 3 Community License
|
|
gptkbp:notableFeature
|
faster inference
better instruction following
improved reasoning
higher accuracy
|
|
gptkbp:openSource
|
partially
|
|
gptkbp:parameter
|
8 billion
|
|
gptkbp:partOf
|
Llama 3 family
|
|
gptkbp:platform
|
gptkb:microprocessor
gptkb:graphics_card
gptkb:TPU
|
|
gptkbp:predecessor
|
gptkb:Llama_2_7B_Chat
|
|
gptkbp:releaseDate
|
April 2024
|
|
gptkbp:size
|
8 billion parameters
|
|
gptkbp:successor
|
gptkb:Llama_2_7B_Chat
|
|
gptkbp:supports
|
code generation
question answering
summarization
reasoning tasks
instruction following
|
|
gptkbp:tokenizer
|
Byte-Pair Encoding (BPE)
custom Llama 3 tokenizer
|
|
gptkbp:trainer
|
Meta-licensed data
reinforcement learning from human feedback
supervised fine-tuning
synthetic data
publicly available data sources
|
|
gptkbp:trainingObjective
|
next token prediction
|
|
gptkbp:bfsParent
|
gptkb:LLaMA_model_family
|
|
gptkbp:bfsLayer
|
8
|