Showing 3 open source projects for "a byte of python"

View related business solutions
  • Managed File Transfer Software Icon
    Managed File Transfer Software

    Products to help you get data where it needs to go—securely and efficiently.

    For too many businesses, complex file transfer needs make it difficult to create, manage and support data flows to and from internal and external systems. Progress® MOVEit® empowers enterprises to take control of their file transfer workflows with solutions that help secure, simplify and centralize data exchanges throughout the organization.
    Learn More
  • More Bookings. Better Experience. Icon
    More Bookings. Better Experience.

    For tour and activity providers

    The all-in-one solution built to help you stay organised and get more bookings with thousands of connections to online travel agencies (OTAs), resellers and suppliers.
    Learn More
  • 1
    minbpe

    minbpe

    Minimal, clean code for the Byte Pair Encoding (BPE) algorithm

    minbpe is a minimal, clean implementation of byte-level Byte Pair Encoding (BPE), the tokenization approach widely used in modern language models. It operates on UTF-8 encoded bytes rather than Unicode characters, which makes it robust to arbitrary text inputs and avoids needing a language-specific character vocabulary. The repository is structured as a teaching-oriented implementation that shows how to train a tokenizer by learning merge rules, then apply those merges to encode text into...
    Downloads: 0 This Week
    Last Update:
    See Project
  • 2
    Tiktoken

    Tiktoken

    tiktoken is a fast BPE tokeniser for use with OpenAI's models

    ...Internally, it includes the core tokenizer logic (often implemented in Rust or efficient lower-level code), APIs for encoding, decoding, and counting tokens, and binding layers to Python (and sometimes other languages) for easy use.
    Downloads: 5 This Week
    Last Update:
    See Project
  • 3
    LaMDA-pytorch

    LaMDA-pytorch

    Open-source pre-training implementation of Google's LaMDA in PyTorch

    Open-source pre-training implementation of Google's LaMDA research paper in PyTorch. The totally not sentient AI. This repository will cover the 2B parameter implementation of the pre-training architecture as that is likely what most can afford to train. You can review Google's latest blog post from 2022 which details LaMDA here. You can also view their previous blog post from 2021 on the model.
    Downloads: 0 This Week
    Last Update:
    See Project
  • Previous
  • You're on page 1
  • Next
MongoDB Logo MongoDB