Amanvir Parhar
@amanvir.bsky.social
building azigy.com • studying @ umd cs • making things amanvir.com • writing venusgirdle.com
I got an LLM to dynamically set its own temperature!
Just wrote a blog post on how I did this and why dynamic temperature adjustment might be a valuable ability for a language model to possess.
Read here: amanvir.com/blog/getting-an-llm-to-set-its-own-temperature
Just wrote a blog post on how I did this and why dynamic temperature adjustment might be a valuable ability for a language model to possess.
Read here: amanvir.com/blog/getting-an-llm-to-set-its-own-temperature
June 24, 2025 at 11:48 PM
I got an LLM to dynamically set its own temperature!
Just wrote a blog post on how I did this and why dynamic temperature adjustment might be a valuable ability for a language model to possess.
Read here: amanvir.com/blog/getting-an-llm-to-set-its-own-temperature
Just wrote a blog post on how I did this and why dynamic temperature adjustment might be a valuable ability for a language model to possess.
Read here: amanvir.com/blog/getting-an-llm-to-set-its-own-temperature
My work was featured in The New Yorker this week!
The "Goings On" section, which focuses on curating cultural touchstones and happenings, linked to my interactive webpage on Earth's obscure islands as part of their annual "Summer Culture Preview."
The "Goings On" section, which focuses on curating cultural touchstones and happenings, linked to my interactive webpage on Earth's obscure islands as part of their annual "Summer Culture Preview."
May 26, 2025 at 10:40 AM
My work was featured in The New Yorker this week!
The "Goings On" section, which focuses on curating cultural touchstones and happenings, linked to my interactive webpage on Earth's obscure islands as part of their annual "Summer Culture Preview."
The "Goings On" section, which focuses on curating cultural touchstones and happenings, linked to my interactive webpage on Earth's obscure islands as part of their annual "Summer Culture Preview."
I created a 3D visualization that shows *every* attention weight matrix within GPT-2 as the model generates tokens!
April 20, 2025 at 5:31 PM
I created a 3D visualization that shows *every* attention weight matrix within GPT-2 as the model generates tokens!
Over the past few days, I worked on creating a text animation effect using Dawkins' weasel, a computer simulation devised by evolutionary biologist Richard Dawkins in the 1980s.
March 18, 2025 at 11:02 AM
Over the past few days, I worked on creating a text animation effect using Dawkins' weasel, a computer simulation devised by evolutionary biologist Richard Dawkins in the 1980s.
I built an ESP32-based DNS sinkhole that stops you from doomscrolling!
This was a really fun project that I wrote a detailed blog post about.
In it, I document my experience building this thing, and I also share what I’ve learned about DNS, networking, and low-level programming along the way ⬇️
This was a really fun project that I wrote a detailed blog post about.
In it, I document my experience building this thing, and I also share what I’ve learned about DNS, networking, and low-level programming along the way ⬇️
February 28, 2025 at 10:35 AM
I built an ESP32-based DNS sinkhole that stops you from doomscrolling!
This was a really fun project that I wrote a detailed blog post about.
In it, I document my experience building this thing, and I also share what I’ve learned about DNS, networking, and low-level programming along the way ⬇️
This was a really fun project that I wrote a detailed blog post about.
In it, I document my experience building this thing, and I also share what I’ve learned about DNS, networking, and low-level programming along the way ⬇️
The model is asked to respond with JSON that contains the indices for the first and last paragraphs of each chapter. Gemini is able to accurately return the index of any given paragraph, because each “paragraph” from the book’s raw text is wrapped with a numbered tag which the model can reference:
February 12, 2025 at 11:33 PM
The model is asked to respond with JSON that contains the indices for the first and last paragraphs of each chapter. Gemini is able to accurately return the index of any given paragraph, because each “paragraph” from the book’s raw text is wrapped with a numbered tag which the model can reference:
The model also performs pretty well at “needle In a haystack”-type tests, so it seemed like the perfect choice! (some tweets by @jeffdean.bsky.social!)
February 12, 2025 at 11:33 PM
The model also performs pretty well at “needle In a haystack”-type tests, so it seemed like the perfect choice! (some tweets by @jeffdean.bsky.social!)
I started thinking about fixing some of these typesetting/formatting issues using an LLM, and I quickly realized that Gemini would be perfect for this project - the 1M+ token context window would allow me to paste the text for an entire book in one shot.
February 12, 2025 at 11:33 PM
I started thinking about fixing some of these typesetting/formatting issues using an LLM, and I quickly realized that Gemini would be perfect for this project - the 1M+ token context window would allow me to paste the text for an entire book in one shot.
This project was born out of a personal need: I was looking for a poetry collection by Robert Frost on gutenberg.org, a repository with thousands of free eBooks, but when I downloaded the ePub for this particular book, I was rather disappointed.
February 12, 2025 at 11:33 PM
This project was born out of a personal need: I was looking for a poetry collection by Robert Frost on gutenberg.org, a repository with thousands of free eBooks, but when I downloaded the ePub for this particular book, I was rather disappointed.
I wrote a script that uses Google’s Gemini-1.5-Flash model to fix typesetting and formatting issues in public domain eBooks from Project Gutenberg.
It’s open source, and it works thanks to Gemini’s insanely long context window (1M+ tokens)!
A thread 🧵 ⬇️
It’s open source, and it works thanks to Gemini’s insanely long context window (1M+ tokens)!
A thread 🧵 ⬇️
February 12, 2025 at 11:33 PM
I wrote a script that uses Google’s Gemini-1.5-Flash model to fix typesetting and formatting issues in public domain eBooks from Project Gutenberg.
It’s open source, and it works thanks to Gemini’s insanely long context window (1M+ tokens)!
A thread 🧵 ⬇️
It’s open source, and it works thanks to Gemini’s insanely long context window (1M+ tokens)!
A thread 🧵 ⬇️