character-level-language-model

This is a diacritization model for Arabic language. This model was built/trained using the Tashkeela: the Arabic diacritization corpus on Kaggle

Official code for Group-Transformer (Scale down Transformer by Grouping Features for a Lightweight Character-level Language Model, COLING-2020).

My journey in AI, documented.

Lyrics Generation:notes: using LSTM , word2vec Analysis and more

Text Article generator using using Character level LSTM network.

Build a character level language model to generate new dinosaur names

Sequence Models coding assignments

A causal intervention framework to learn robust and interpretable character representations inside subword-based language models

中国诗词的模型训练，一句话就能变为一首诗

Configurable character-level transformer training suite with built-in mechanistic interpretability toolkit — scale to 150M+ parameters and beyond, no ceilings, only hardware limits. Inspect attention weights, hidden states, and head specialisation across all layers. Documented circuit findings included.

In this project, I worked with a small corpus consisting of simple sentences. I tokenized the words using n-grams from the NLTK library and performed word-level and character-level one-hot encoding. Additionally, I utilized the Keras Tokenizer to tokenize the sentences and implemented word embedding using the Embedding layer. For sentiment analysis

Notebooks of programming assignments of Sequence Models course of deeplearning.ai on coursera in May-2020

An implementation of "Character-level Convolutional Networks for Text Classification" in Tensorflow. See https://arxiv.org/pdf/1509.01626.pdf.

Saryu: a recurrent language model whose state is moved by input-dependent Householder reflections. Open architecture research from India — constant memory per token, an exact parallel kernel, and two technical reports: the architecture, and what the same transport does when it fails (it collapses onto exact group quotients).

It aims to write new sentences by learning character units sentences using RNN. As training data, a collection of Shakespeare's novels was used.

Character-level GPT trained on Truyện Kiều (Nguyễn Du), based on Karpathy's nanoGPT

This repository contains the code and PLODv2 dataset to train character-level language models (CLM) for abbreviation and long-form detection released with our LREC-COLING 2024 publication

Annotated study fork of Karpathy nanoGPT — GPT-2 training from scratch with extended notes on causal self-attention, positional encoding, layer norm placement, and efficient fine-tuning on custom datasets.

This repository demonstrates a minimal character-level recurrent neural network language model built with PyTorch. It trains on the Tiny Shakespeare corpus popularized by Andrej Karpathy, providing both the original Google Colab notebook and an exported Python script for reference or interactive execution.

Minimal PyTorch implementations of Bigram and GPT language models from scratch, with KV caching.