normalcereal
We like to help computers talk to, with, and overall "understand" people.
Parker, Colorado
Pinned Repositories
AIDungeon
Infinite adventures await!
Amaze-Scraper
Amazon product search and walmart.com price comparation.
bobbyjo-solutions
clstrfck
clstrfck -- a simple encoder that reduces textual english word length using a consistent, static, ruleset. Originally created to reduce the number of tokens used in text generation so as to increase information density in text generation priming and prompting, whilst outputting text that decodes to a significantly greater length than previously possible. First used for fine-tuning in language models, then used for prompting, assuming the language model can -- for the most part -- connect the relativistic linguistic dots for the shorter token lengths in the vocabulary.
gpt-2-simple-XR
Forking toward an extended release version for myself and maybe others... to better suit my current and future needs...in ways that don't match goals of origin package. Original description: Python package to easily retrain OpenAI's GPT-2 text-generating model on new texts
jina
An easier way to build neural search in the cloud
jobtalk
resume
resume repo
normalcereal's Repositories
normalcereal/AIDungeon
Infinite adventures await!
normalcereal/clstrfck
clstrfck -- a simple encoder that reduces textual english word length using a consistent, static, ruleset. Originally created to reduce the number of tokens used in text generation so as to increase information density in text generation priming and prompting, whilst outputting text that decodes to a significantly greater length than previously possible. First used for fine-tuning in language models, then used for prompting, assuming the language model can -- for the most part -- connect the relativistic linguistic dots for the shorter token lengths in the vocabulary.
normalcereal/gpt-2-simple-XR
Forking toward an extended release version for myself and maybe others... to better suit my current and future needs...in ways that don't match goals of origin package. Original description: Python package to easily retrain OpenAI's GPT-2 text-generating model on new texts
normalcereal/jina
An easier way to build neural search in the cloud
normalcereal/jobtalk
normalcereal/resume
resume repo
normalcereal/Amaze-Scraper
Amazon product search and walmart.com price comparation.
normalcereal/bobbyjo-solutions
normalcereal/ceMEnt
core entity mirror entity nom...tensor... • paper...
normalcereal/MLTS
Machine Learning Toolkit for SEO
normalcereal/recovery_from_covid_scare
normalcereal/Urbanist
Urbanist is a low-contrast, geometric sans-serif inspired by Modernist design and typography.
normalcereal/vuetify
🐉 Material Component Framework for Vue