Posts
Transformers FLOPS and memory usage
February 26, 2025 - An in-depth breakdown with an example of how much GPU memory is required to run inference using a Transformer
February 15, 2025 - What happens under the hood when we overwrite a variable but that is not an in-place operation?
January 12, 2025 - Define ResNet as detailed in the paper, train it on CIFAR-10 and reproduce the classification results on CIFAR-10
Retrieval Augmented Generation
December 24, 2024 - Fine-tuning BERT to encode sentences and GPT-2 for question answering. Can I generate meaningful outputs using these relatively small models?
Sentiment Classification on IMDb dataset
November 30, 2024 - Use LoRA to fine-tune GPT-2 and BERT for sentiment classification.