#
Lists (31)
agentset
6 repositories
arabic_nlp
12 repositories
ATS
4 repositories
backend_mastering
3 repositories
chatbot
Rasa, botpress, api and general frameworks for creating chatbots1 repository
Code Reading list
This list is a code repos which i want to read in my free time and check2 repositories
cool_try_to_build
5 repositories
distil_llm
2 repositories
DSPy
1 repository
embedding_world
3 repositories
explore it with curisoty
10 repositories
frontend engineering
4 repositories
general nlp
15 repositories
gnn
1 repository
graph
5 repositories
intraoral
3 repositories
LLMs
2 repositories
must_master
3 repositories
OCR
1 repository
qdrant
2 repositories
Recs
1 repository
research
2 repositories
robotics
1 repository
signal_processing
2 repositories
sports
2 repositories
substacks
1 repository
super_user
1 repository
tokenization
3 repositories
torch
5 repositories
usul
2 repositories
vector_databases
14 repositories
Starred repositories
Trainable embedding transformation for confidence estimation, feature extraction, explainability and conversion from dense to sparse.
Understand and test language model architectures on synthetic tasks.
VS Code extension for LLM-assisted code/text completion
Vim plugin for LLM-assisted code/text completion
Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends
aider is AI pair programming in your terminal
Think DSP: Digital Signal Processing in Python, by Allen B. Downey.
🦉 Your Junk-Free, Personalized Information and Podcasts.
A framework for evaluating semantic search across custom datasets, metrics, and embedding backends.
Simple UI for debugging correlations of text embeddings
LLM-driven automated knowledge graph construction from text using DSPy and Neo4j.
In-browser tool to explore semantic similarity segmenting strategies by generating and querying overlapping text segments using Transformers.js.
Download free and premium Substack posts, saving them as Markdown files. Also generates HTML interfaces to allow you to browse and sort the markdown files for each author.
This is our own implementation of 'Layer Selective Rank Reduction'