Contents
attention
Performer vs Softmax — From Kernel View to Fair Speed Tests • Aug 30, 2025
Attention Paths and Rank Collapse Part 1 • Mar 20, 2025
backpropagation
Making FSDP2 and QLoRA Work Together: Training Llama 8B on 2 GPUs • Dec 15, 2025
Memory-Efficient Backprop: Solving Unsloth's Challenge E • Dec 5, 2025
cnn
Do we need downsampling? • May 2, 2020
convolution
im2col • Dec 23, 2023
deeplearning
From Zero to 1.55x: Writing My First Triton Kernel for NF4 Dequantization • Dec 21, 2025
Making FSDP2 and QLoRA Work Together: Training Llama 8B on 2 GPUs • Dec 15, 2025
Memory-Efficient Backprop: Solving Unsloth's Challenge E • Dec 5, 2025
Performer vs Softmax — From Kernel View to Fair Speed Tests • Aug 30, 2025
Indic CLIP Multimodal Understanding for Indic Languages • Apr 26, 2025
Creating a Maze Solver using Pix2Pix • Mar 29, 2025
Why Batch Size Matters - The Surprising Difference Between Batch Training and Averaging • Mar 21, 2025
Attention Paths and Rank Collapse Part 1 • Mar 20, 2025
Unveiling Position Encoding in Transformers - From Absolute to Relative with RoPE • Mar 1, 2025
Understanding Transformer Positional Encodings - A Mathematical Deep Dive • Feb 22, 2025
Understanding and Preventing Collapse in Self-Supervised Learning A Deep Dive into BYOL • Feb 4, 2025
Gradient Clipping and Adaptive Learning Rates • Jan 12, 2025
Deep-Contextualized Embeddings ( ELMO ) • Dec 17, 2024
Tight-fisted Optimizer ( Tiger ) • Nov 19, 2024
im2col • Dec 23, 2023
Residual Learning • Sep 19, 2022
AEDA ( An Easier Data Augmentation Technique for Text Classification ) • Jan 31, 2022
Temporal Convolution Networks • Dec 22, 2021
Do we need downsampling? • May 2, 2020
deepseek
DSpark: Confidence-Scheduled Speculative Decoding • Jun 29, 2026
fast.ai
Understanding and Preventing Collapse in Self-Supervised Learning A Deep Dive into BYOL • Feb 4, 2025
Gradient Clipping and Adaptive Learning Rates • Jan 12, 2025
Deep-Contextualized Embeddings ( ELMO ) • Dec 17, 2024
fastai
Performer vs Softmax — From Kernel View to Fair Speed Tests • Aug 30, 2025
Indic CLIP Multimodal Understanding for Indic Languages • Apr 26, 2025
AEDA ( An Easier Data Augmentation Technique for Text Classification ) • Jan 31, 2022
Temporal Convolution Networks • Dec 22, 2021
Do we need downsampling? • May 2, 2020
gan
Creating a Maze Solver using Pix2Pix • Mar 29, 2025
gpu
From Zero to 1.55x: Writing My First Triton Kernel for NF4 Dequantization • Dec 21, 2025
inference
DSpark: Confidence-Scheduled Speculative Decoding • Jun 29, 2026
llm
DSpark: Confidence-Scheduled Speculative Decoding • Jun 29, 2026
Performer vs Softmax — From Kernel View to Fair Speed Tests • Aug 30, 2025
Attention Paths and Rank Collapse Part 1 • Mar 20, 2025
Unveiling Position Encoding in Transformers - From Absolute to Relative with RoPE • Mar 1, 2025
Understanding Transformer Positional Encodings - A Mathematical Deep Dive • Feb 22, 2025
math
Creating a Maze Solver using Pix2Pix • Mar 29, 2025
Why Batch Size Matters - The Surprising Difference Between Batch Training and Averaging • Mar 21, 2025
Attention Paths and Rank Collapse Part 1 • Mar 20, 2025
Unveiling Position Encoding in Transformers - From Absolute to Relative with RoPE • Mar 1, 2025
Understanding Transformer Positional Encodings - A Mathematical Deep Dive • Feb 22, 2025
Understanding and Preventing Collapse in Self-Supervised Learning A Deep Dive into BYOL • Feb 4, 2025
Gradient Clipping and Adaptive Learning Rates • Jan 12, 2025
Deep-Contextualized Embeddings ( ELMO ) • Dec 17, 2024
Tight-fisted Optimizer ( Tiger ) • Nov 19, 2024
im2col • Dec 23, 2023
Residual Learning • Sep 19, 2022
AEDA ( An Easier Data Augmentation Technique for Text Classification ) • Jan 31, 2022
Temporal Convolution Networks • Dec 22, 2021
multimodal
Indic CLIP Multimodal Understanding for Indic Languages • Apr 26, 2025
nlp
AEDA ( An Easier Data Augmentation Technique for Text Classification ) • Jan 31, 2022
optimization
Gradient Clipping and Adaptive Learning Rates • Jan 12, 2025
Deep-Contextualized Embeddings ( ELMO ) • Dec 17, 2024
Tight-fisted Optimizer ( Tiger ) • Nov 19, 2024
self-supervised-learning
Understanding and Preventing Collapse in Self-Supervised Learning A Deep Dive into BYOL • Feb 4, 2025
sequencemodelling
Temporal Convolution Networks • Dec 22, 2021
speculative-decoding
DSpark: Confidence-Scheduled Speculative Decoding • Jun 29, 2026
systems
DSpark: Confidence-Scheduled Speculative Decoding • Jun 29, 2026
Attention Paths and Rank Collapse Part 1 • Mar 20, 2025
Unveiling Position Encoding in Transformers - From Absolute to Relative with RoPE • Mar 1, 2025
Understanding Transformer Positional Encodings - A Mathematical Deep Dive • Feb 22, 2025
triton
From Zero to 1.55x: Writing My First Triton Kernel for NF4 Dequantization • Dec 21, 2025
tsai
Temporal Convolution Networks • Dec 22, 2021
unsloth
From Zero to 1.55x: Writing My First Triton Kernel for NF4 Dequantization • Dec 21, 2025
Making FSDP2 and QLoRA Work Together: Training Llama 8B on 2 GPUs • Dec 15, 2025
Memory-Efficient Backprop: Solving Unsloth's Challenge E • Dec 5, 2025