Sasikanth Kotti PRO

Sasikanth

https://ksasi.github.io/

AI & ML interests

Computer Vision, Adversarial Robustness, Machine Unlearning, Model Merging, Self Supervised Learning

Recent Activity

upvoted a collection 10 days ago

Sparse Auto-Encoders (SAEs) for Mechanistic Interpretability

liked a Space 19 days ago

OpenEvals/evaluation-guidebook

updated a collection about 1 month ago

Image_Classifiers

View all activity

Organizations

upvoted a collection 10 days ago

Sparse Auto-Encoders (SAEs) for Mechanistic Interpretability

Collection

A compilation of sparse auto-encoders trained on large language models. • 37 items • Updated 12 days ago • 16

liked a Space 19 days ago

Evaluation Guidebook

📝

217

Display benchmark evaluation data for LLMs

updated a collection about 1 month ago

Image_Classifiers

Collection

2 items • Updated Nov 22

upvoted an article about 1 month ago

Article

Easily Build and Share ROCm Kernels with Hugging Face

Nov 17

•

updated a collection 2 months ago

Training Tricks

Collection

3 items • Updated Oct 20

liked a Space 2 months ago

Gpu Tflop Finder

😻

Get the TFLOPs for your GPU quickly

reacted to burtenshaw's post with 🤗❤️🔥 3 months ago

Post

6000

Smol course has a distinctive approach to teaching post-training, so I'm posting about how it’s different to other post-training courses, including the llm course that’s already available.

In short, the smol course is just more direct that any of the other course, and intended for semi-pro post trainers.

- It’s a minimal set of instructions on the core parts.
- It’s intended to bootstrap real projects you're working on.
- The material handsover to existing documentation for details
- Likewise, it handsover to the LLM course for basics.
- Assessment is based on a leaderboard, without reading all the material.

To start the smol course, follow here: