|
gptkbp:instance_of
|
gptkb:language
|
|
gptkbp:api
|
available
|
|
gptkbp:application
|
natural language processing
|
|
gptkbp:architecture
|
gptkb:Transformers
|
|
gptkbp:collaboration
|
gptkb:open_AI
gptkb:Stanford_University
gptkb:University_of_Washington
gptkb:Carnegie_Mellon_University
gptkb:Google_Research
research institutions
|
|
gptkbp:community_support
|
active
|
|
gptkbp:contribution
|
research papers
tutorials
workshops
code repositories
open datasets
|
|
gptkbp:developed_by
|
gptkb:Meta
|
|
gptkbp:evaluates
|
perplexity
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
LLa MA
|
|
gptkbp:is_open_source
|
gptkb:true
|
|
gptkbp:language_support
|
multiple languages
|
|
gptkbp:latest_version
|
gptkb:LLa_MA_2
|
|
gptkbp:license
|
Research License
|
|
gptkbp:model
|
7 B, 13 B, 30 B, 65 B parameters
|
|
gptkbp:parameter_efficiency
|
high
|
|
gptkbp:performance
|
gptkb:GLUE
gptkb:Super_GLUE
gptkb:SQu_AD
gptkb:Winograd_Schema_Challenge
state-of-the-art
Hella Swag
|
|
gptkbp:predecessor
|
gptkb:GPT-3
|
|
gptkbp:project
|
gptkb:LLa_MA_4
improved efficiency
broader applications
better fine-tuning methods
enhanced multilingual capabilities
|
|
gptkbp:provides_information_on
|
text data
|
|
gptkbp:release_date
|
gptkb:2023
|
|
gptkbp:research_focus
|
AI safety
AI ethics
AI scalability
AI bias
AI interpretability
|
|
gptkbp:successor
|
gptkb:LLa_MA_3
|
|
gptkbp:training
|
self-supervised learning
|
|
gptkbp:training_compute
|
high
|
|
gptkbp:training_programs
|
gptkb:Py_Torch
|
|
gptkbp:tuning
|
possible
|
|
gptkbp:use_case
|
gptkb:translator
question answering
chatbots
text generation
summarization
|
|
gptkbp:user_base
|
gptkb:students
gptkb:developers
gptkb:researchers
companies
|
|
gptkbp:bfsParent
|
gptkb:Hugging_Face
|
|
gptkbp:bfsLayer
|
5
|