you will know more than 90% of ml enthusiasts just by reading these :)

yo this book is over chat, genuinely good read do consider this after homl. im thinking abt going with AI eng as the next one, also i have to make 2 or 3 projects man plus its been a while we haven't read a paper so we'll start that too after diwali we start with new book yay :)

ml goals for the day chat (chill day) - will explore fineweb dataset & there is a blog on 2M+ neural embeddings based web search engine - just for curiosity we'll explore that - will update my resume today, also its diwali bro we'll chill what ru guys doing today??

how tf will you debug this??

you have a life but are you living :)

ok so final chapter to go, some fck happened but please book keep things easy for me. beside the book the git repo of llm from scratch is really good like bonus materials are also very good. ik i'll revisit this repo alot of times :) repo link 👇

me after getting a remote job :)

ml goals for the day chat - finish build llm from scratch there are around 50 pages left - plan is that before diwali we will train our gpt model on fineweb - no maths for today cause ml will take most of the time - need to update my resume what u guys doing, diwali plans??

im between stage 1 and 2 :)

"Some people are actually living like this. is this the life you want"

fckk its 4: 00 am around 50 pages are still left, plan is to end to this book today so we can work on training our gpt model on fineweb dataset cause after diwali i want to start with a new book. lets see how much we can do tmrw also didn't read maths, this shit is more imp :)

we set clear goals for the day, they are non negotiable man. we lock in until the goals are achieved, rn big portion of ml is left & then even if i read 2 or 3 pages of maths i'll be happy cause consistency is imp babu :)

sorry i will learn cpp very soon :)

i never made notes in ml cause its like evrything is in my head, but like atm nag is lil blurry so i'll just google that but in math things are so messy with so many notations, symbols that u have to constantly recall small details. so if you're starting maths please make notes✌️

ml goals for the day chat - will finish with fine tuning for classification ch 6 - making notes in maths was a good decision no cap - will go through proof of cauchy schwarz then orthogonality - in maths we're going slow cause it can be too much confusing what abt u chat :)

“Life is a journey, and if you fall in love with the journey, you will be in love forever.”

yup so done with pretraining on unlabeled dataset but this was on a small text file. very soon we will train this on fine web dataset which is huge, now only two chapters to go which are easy af atleast this is what i think. this book is really great tbh after homl :)

"the most certain way to succeed is always to try just one more time"

i really love GPUs :)

quite funny that im in 5th sem & i haven't got a single internship even though im good in ml ofc not cracked. reason is that i haven't built ml projects not even shitty ones & didn't even bother to apply cause now ik how much is necessary & i think around 6th sem we'll apply :)

ml goals for the day chat - pretraining model on unlabeled dataset - a lil change in plan there are 3 chapters left so we'll cover them till 15 & then train our gpt-2 model on fineweb dataset - maths we'll start with generated norm, cauchy-schwarz inequality what abt u chat :)