Implementation of GPT from scratch. Design to be lightweight and easy to modify.
A project implementing and comparing from-scratch RNN (GRU) and Transformer models for next-word prediction on a Persian Wikipedia dataset.
Seq2Seq model implemented with pytorch, using temperature sampling, top-k sampling, top-p sampling / nucleus sampling, and beam search.
Pipelining, and implementing RAG, and evaluating comparative performance across enhancements
Conversational agent using MultiWOZ dataset and decoding methods as beam search, and top-k sampling
A serie of notebooks exploring GPT-Neo
Demonstrates how GenAI parameters like Temperature, top-K and top-P work
Add a description, image, and links to the top-k-sampling topic page so that developers can more easily learn about it.
To associate your repository with the top-k-sampling topic, visit your repo's landing page and select "manage topics."