Latest Tech News

Stay updated with the latest in technology, AI, cybersecurity, and more

Filtered by: torch Clear Filter

Show HN: Learn LLMs LeetCode Style

news.ycombinator.com Unknown 2025-07-28 23:03:13

TorchLeet is broken into two sets of questions: Question Set: A collection of PyTorch practice problems, ranging from basic to hard, designed to enhance your skills in deep learning and PyTorch. LLM Set: A new set of questions focused on understanding and implementing Large Language Models (LLMs) from scratch, including attention mechanisms, embeddings, and more. Note Avoid using GPT. Try to solve these problems on your own. The goal is to learn and understand PyTorch concepts deeply. Table o

Topics: problems pytorch question questions set
Read More
Shop Amazon

AI leadership development platform Praxis Labs sells to Torch

techcrunch.com Dominic-Madori Davis 2025-07-29 13:00:00

Praxis Labs, a learning development platform, announced its acquisition on Thursday for an undisclosed amount by the leadership and coaching enterprise Torch. “As a small company with fewer than 20 people serving companies as large as Amazon, we knew we needed to build powerful partnerships, across product and go-to-market, to reach more companies,” co-founder and CEO of Praxis Labs, Elise Smith, told TechCrunch about the reason for the sale. She and her co-founder, Heather Shen, met Torch CEO

Topics: ai heather praxis torch ventures
Read More
Shop Amazon

Showh HN: Microjax – JAX in two classes and six functions

news.ycombinator.com Unknown 2025-08-08 20:41:55

Microjax: JAX in two classes and six functions or Read on Github (I recommend actually running the notebook, either on your own computer or Colab). This is inspired by Andrej Karpathy's Micrograd, a PyTorch-like library in about 150 lines of code. Despite PyTorch's popularity, I prefer the way JAX works because it a more functional style. This tutorial borrows heavily from Matthew J Johnson's great 2017 presentation on the predecessor to JAX, autograd: Video / Slides / Code. My main contribut

Topics: autograd code jax notebook pytorch
Read More
Shop Amazon

Showh HN: Microjax - Jax in two classes and six functions

news.ycombinator.com Unknown 2025-08-09 01:41:55

Microjax: JAX in two classes and six functions or Read on Github (I recommend actually running the notebook, either on your own computer or Colab). This is inspired by Andrej Karpathy's Micrograd, a PyTorch-like library in about 150 lines of code. Despite PyTorch's popularity, I prefer the way JAX works because it a more functional style. This tutorial borrows heavily from Matthew J Johnson's great 2017 presentation on the predecessor to JAX, autograd: Video / Slides / Code. My main contribut

Topics: autograd code jax notebook pytorch
Read More
Shop Amazon

Fault Tolerant Llama training – PyTorch blog

news.ycombinator.com Tristan Rice 2025-08-19 11:30:02

Collaborators: Less Wright, Howard Huang, Chien-Chin Huang, Crusoe: Martin Cala, Ethan Petersen tl;dr: we used torchft and torchtitan to train a model in a real-world environment with extreme synthetic failure rates to prove reliability and correctness of fault tolerant training Training loss across 1200 failures with no checkpoints. NOTE: Each small spike is a non-participating worker recovering which affects the metrics but not the model Introduction We want to demonstrate torchft in wo

Topics: group replica step torchft training
Read More
Shop Amazon

PyTorch Reshaping with None

news.ycombinator.com Unknown 2025-08-24 19:03:45

PyTorch Reshaping with None Currently I am learning attention mechanism from Dive into Deep Learning book. In the book I see following implementation in masked softmax: def sequence_mask (X, valid_len, value = - 1e6 ): """ X is 2D array (number_of_points, maxlen), valid_len is 1D array (number_of_points)""" max_len = X . size( 1 ) mask = torch . arange(max_len, dtype = torch . float32, device = X . device)[ None , :] < valid_len[:, None ] X[ ~ mask] = value return X In sequential data process

Topics: device mask max_len torch valid_len
Read More
Shop Amazon

DeepDive in everything of Llama3: revealing detailed insights and implementation

news.ycombinator.com Unknown 2025-06-01 13:57:13

[ View in English | 中文版文档点这里 ] This project is an enhanced version based on naklecha/llama3-from-scratch. It has been comprehensively improved and optimized on the basis of the original project, aiming to help everyone more easily understand and master the implementation principle and the detailed reasoning process of the Llama3 model. Thanks to the contributions of the original author :) The following are the core improvements of this project: Structural Optimization The presentation se

Topics: attention inf token tokens torch
Read More
Shop Amazon
Trending Topics
Hot Now
ai 1456 apple 1083 new 885 content 836 google 668
Popular
zdnet 643 amazon 643 like 633 does 524 reviews 517 company 512 editorial 503 day 489 app 476 prime 445
Emerging
said 439 data 438 samsung 392 phone 388 pro 377 game 376 just 367 android 349 deals 336 galaxy 320
About GoKawiil

GoKawiil is a project by nerdhub.co that curates technology news from trusted sources. We built this site to provide a cleaner, more mobile-friendly experience without intrusive ads or heavy JavaScript.

Privacy

Your privacy matters. GoKawiil doesn't use Google Analytics or Facebook Pixel. The only tracking occurs through affiliate links to Amazon.

Advertising

Interested in advertising? Contact us at [email protected]