Anton Bacaj

fine-tune-mistral. Fine-tune mistral-7B on 3090s, a100s, h100s

awesome-transformers. A curated list of awesome transformer models.

mpt-30B-inference. Run inference on MPT-30B using CPU

code-eval. Run evaluation on LLMs using human-eval benchmark

chatgpt-backup. Single client side script to backup your entire ChatGPT conversation history

replit-3B-inference. Run inference on replit-3B code instruct model using CPU

openhermes-function-calling. Jupyter Notebook

transformers. Understanding large language models

train-with-fsdp. Python

unofficial-chatgpt-api. Node.js client for the chatgpt API. No third party dependencies.

react-simple-filter-sort-ecommerce. Repo used in this video: https://youtu.be/c3WSziz_u_o

transformers-docker. Run, build, test transformer models using docker

resolutejs. Finally get to retry during a Promise operation (works in modern browsers as well as nodejs), zero dependencies.

node-selenium-starter. A starter project for automated browser testing using Node.js.

adept-inference-local-3090. Forked version of the adept repo deleting some distributed code

ctransformers. Python bindings for the Transformer models implemented in C/C++ using GGML library.

electron-easy-ts. Build electron apps with typescript the easy way.

concise.css. Lightweight, SASS-based, OOCSS principles, bloat-free (we promise).

local-code-eval. Jupyter Notebook