Statements (28)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:large_language_model
|
| gptkbp:architecture |
gptkb:transformation
|
| gptkbp:availableOn |
gptkb:Hugging_Face
TII website |
| gptkbp:countryOfOrigin |
gptkb:United_Arab_Emirates
|
| gptkbp:developedBy |
gptkb:Technology_Innovation_Institute
|
| gptkbp:fineTunedWith |
true
|
| gptkbp:hardware |
A100 GPUs recommended
|
| gptkbp:hasModel |
decoder-only
|
| gptkbp:intendedUse |
gptkb:research
commercial |
| gptkbp:language |
English
|
| gptkbp:license |
Apache 2.0
|
| gptkbp:notableFor |
first open-source 40B parameter LLM
high performance on open LLM benchmarks |
| gptkbp:openSource |
true
|
| gptkbp:parameter |
40 billion
|
| gptkbp:pretrained |
true
|
| gptkbp:releaseDate |
May 2023
|
| gptkbp:supports |
code generation
question answering text generation text summarization |
| gptkbp:trainer |
1 trillion tokens
|
| gptkbp:trainingDataSource |
RefinedWeb dataset
|
| gptkbp:bfsParent |
gptkb:Falcon_LLM
|
| gptkbp:bfsLayer |
8
|
| https://www.w3.org/2000/01/rdf-schema#label |
Falcon-40B
|