# @awomanindatascience on Instagram

- **Type:** Video
- **Original URL:** https://www.instagram.com/p/DTmZMGtk_SJ
- **Gondola URL:** https://gondola.cc/posts/61230165-awomanindatascience-instagram
- **Thumbnail:** https://img.gondola.cc/tr:w-,h-,fo-auto/postThumbnails/772a008c05.jpg
- **Posted:** 2026-01-17T05:24:43.000+00:00
- **Account Owner:** Chinar Arora (@awomanindatascience) — https://gondola.cc/awomanindatascience

## Caption

Happy Friday! Finally got some 🌞 in Seattle!
For today’s Day 16 of Building an LLM from scratch, I 
Implemented#deeplearning #generativeai #womenwhocode #largelanguagemodels #womenintech a full Transformer block:
causal multi-head self-attention → feed-forward network, wrapped with LayerNorm, residual connections, and dropout.
Input and output shapes stay the same, but representations become context-enriched at every layer.
This is the core unit that scales depth in transformers.

## Stats

- **Views:** 0
- **Likes:** 158
- **Shares:** 0
- **Comments:** 7

## Tags

generativeai, womenintech, largelanguagemodels, deeplearning, womenwhocode

---
Copyright (c) Gondola