This is the Linux app named LLMs-from-scratch whose latest release can be downloaded as LLMs-from-scratchsourcecode.tar.gz. It can be run online in the free hosting provider OnWorks for workstations.
Download and run online this app named LLMs-from-scratch with OnWorks for free.
Volg deze instructies om deze app uit te voeren:
- 1. Download deze applicatie op uw pc.
- 2. Voer in onze bestandsbeheerder https://www.onworks.net/myfiles.php?username=XXXXX in met de gebruikersnaam die u wilt.
- 3. Upload deze applicatie in zo'n bestandsbeheerder.
- 4. Start de OnWorks Linux online of Windows online emulator of MACOS online emulator vanaf deze website.
- 5. Ga vanuit het OnWorks Linux-besturingssysteem dat u zojuist hebt gestart naar onze bestandsbeheerder https://www.onworks.net/myfiles.php?username=XXXXX met de gewenste gebruikersnaam.
- 6. Download de applicatie, installeer hem en voer hem uit.
SCREENSHOTS
Ad
LLMs-from-scratch
PRODUCTBESCHRIJVING
LLMs-from-scratch is an educational codebase that walks through implementing modern large-language-model components step by step. It emphasizes building blocks—tokenization, embeddings, attention, feed-forward layers, normalization, and training loops—so learners understand not just how to use a model but how it works internally. The repository favors clear Python and NumPy or PyTorch implementations that can be run and modified without heavyweight frameworks obscuring the logic. Chapters and notebooks progress from tiny toy models to more capable transformer stacks, including sampling strategies and evaluation hooks. The focus is on readability, correctness, and experimentation, making it ideal for students and practitioners transitioning from theory to working systems. By the end, you have a grounded sense of how data pipelines, optimization, and inference interact to produce fluent text.
Kenmerken
- Stepwise implementations of tokenizer, attention, and transformer blocks
- Clear Python notebooks and scripts designed for learning and tinkering
- Training and sampling loops that expose the full data and compute flow
- Explorations of scaling choices, regularization, and evaluation metrics
- Minimal dependencies to keep the math and code transparent
- Serves as a foundation for extending to larger models and custom datasets
Programmeertaal
Python
Categorieën
This is an application that can also be fetched from https://sourceforge.net/projects/llms-from-scratch.mirror/. It has been hosted in OnWorks in order to be run online in an easiest way from one of our free Operative Systems.
