Statements (51)
Predicate | Object |
---|---|
gptkbp:instanceOf |
large language model
|
gptkbp:architecture |
gptkb:transformation
|
gptkbp:availableOn |
gptkb:Meta_AI_website
Hugging Face (with restrictions) |
gptkbp:citation |
gptkb:Touvron_et_al.,_2023
|
gptkbp:developedBy |
gptkb:Meta_AI
|
gptkbp:firstReleased |
2023
|
https://www.w3.org/2000/01/rdf-schema#label |
Llama model
|
gptkbp:input |
gptkb:text
|
gptkbp:language |
English
|
gptkbp:license |
custom Meta license
|
gptkbp:notableEvent |
Llama 2 released July 2023
Llama 3 released April 2024 weights leaked in March 2023 |
gptkbp:notableFeature |
scalable architecture
chat fine-tuning (in later versions) efficient training instruction tuning (in later versions) multi-lingual support (in later versions) open weights (with restrictions) |
gptkbp:notableFor |
AI research
open-source AI community |
gptkbp:notableRelease |
gptkb:Llama_2
gptkb:Llama_3 |
gptkbp:notableUser |
gptkb:researchers
academia developers companies |
gptkbp:openSource |
partially
|
gptkbp:output |
gptkb:text
|
gptkbp:parameter |
13B
33B 65B 7B |
gptkbp:predecessor |
nan
|
gptkbp:purpose |
natural language processing
|
gptkbp:relatedTo |
gptkb:Falcon_LLM
gptkb:GPT-3 gptkb:PaLM gptkb:MPT gptkb:OpenLLaMA |
gptkbp:successor |
gptkb:Llama_2
|
gptkbp:trainer |
supervised learning
publicly available datasets reinforcement learning from human feedback (in later versions) |
gptkbp:uses |
chatbot
question answering summarization text generation |
gptkbp:bfsParent |
gptkb:artificial_intelligence
|
gptkbp:bfsLayer |
4
|