https://github.com/burcgokden/coulgat-graph-attention-interpretability

Implementation of CoulGAT framework and screened Coulomb attention mechanism.

artificial-intelligence attention deep-learning density-functional-theory graph-attention-networks graph-neural-network graphs interpretability machine-learning machine-learning-algorithms neural-network python quantum-chemistry regularization tensorflow

Last synced: 16 May 2025

https://github.com/burcgokden/llm-from-power-law-decoder-representations

Implementation of PLDR-LLM: Large Language Model from Power Law Decoder Representations

dag deep-learning directed-acyclic-graph keras large-language-models llm machine-learning natural-language-processing nlp pldr-llm power-law-decoder-representations power-law-graph-attention tensorflow transformer

Last synced: 12 Feb 2025

https://github.com/burcgokden/bert-subword-tokenizer-wrapper

A framework for generating subword vocabulary from a tensorflow dataset and building custom BERT tokenizer models.

bert berttokenizer deep-learning machine-learning machine-translation subword tensorflow tensorflow-text vocabulary-builder wordpiece

Last synced: 14 Mar 2025

https://github.com/burcgokden/a-robust-deep-cnn-model-using-numpy

Numerically stable implementation of a CNN Model from scratch using Numpy. Model trained on MNIST data.

Last synced: 14 Mar 2025

https://github.com/burcgokden/power-law-graph-transformer

Implementation of Power Law Graph Transformer for Machine Translation and Representation Learning.

attention deep-learning graph-neural-network graph-transformer machine-learning machine-learning-algorithms machine-learning-model machine-translation neural-network plga plgt power-law power-law-graph-transformer representation-learning residual-networks tensorflow tensorflow-text transformer

Last synced: 14 Mar 2025

https://github.com/burcgokden/sentencepiece-tokenizer-wrapper-for-pldr-llm-kvg-cache

SentencePiece Tokenizer Wrapper implementation for PLDR-LLM with KV cache and G-cache

bpe deep-learning large-language-models llm machine-learning natural-language-processing pldr-llm pytorch sentencepiece tokenizer transformer unigram

Last synced: 21 Feb 2025

https://github.com/burcgokden/sdpa-transformer-wrapper

A framework for training and evaluating a transformer with scaled dot product attention on a tensorflow dataset.

attention deep-learning embeddings machine-translation natural-langugage-processing neural-network nlp nmt scaled-dot-product smt tensorflow tensorflow-datasets tensorflow-text transformer

Last synced: 14 Mar 2025

https://github.com/burcgokden/sentencepiece-tokenizer-wrapper-for-pldr-llm

A framework for building Sentencepiece tokenizer from a dataset

bpe deep-learning keras large-language-models llm machine-learning natural-language-processing pldr-llm sentencepiece tensorflow tokenizer transformer unigram

Last synced: 06 Apr 2025

https://github.com/burcgokden/pldr-llm-with-kvg-cache

Implementation of PLDR-LLM with KV-cache and G-cache in Pytorch for the paper titled "PLDR-LLMs Learn A Generalizable Tensor Operator That Can Replace Its Own Deep Neural Net At Inference"

llm pldr-llm pytorch

Last synced: 21 Feb 2025