Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Batched quantization #516

Merged
merged 21 commits into from
Jul 2, 2024
Merged

Batched quantization #516

merged 21 commits into from
Jul 2, 2024

Commits on Apr 6, 2024

  1. Batched quantization

    @casper-hansen
    casper-hansen committed Apr 6, 2024
    Configuration menu
    Copy the full SHA
    fd3a9d4 View commit details
    Browse the repository at this point in the history

Commits on Jun 21, 2024

  1. Merge branch 'main' into batched_quantization

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    6ff2b64 View commit details
    Browse the repository at this point in the history

  2. Enable long context calibration data

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    e53840d View commit details
    Browse the repository at this point in the history

  3. Fix reference bug

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    6875a25 View commit details
    Browse the repository at this point in the history

  4. Fix not selecting right output

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    1e89762 View commit details
    Browse the repository at this point in the history

  5. Add max_calib_samples and max_calib_seq_len

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    08d75cc View commit details
    Browse the repository at this point in the history

  6. Potentially handle overflow

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    bc37fda View commit details
    Browse the repository at this point in the history

  7. Improve naming and docs

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    edefcab View commit details
    Browse the repository at this point in the history

  8. Chunked loss computation

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    a0b3ac7 View commit details
    Browse the repository at this point in the history

  9. Long-context doc on Cosmopedia

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    0006e98 View commit details
    Browse the repository at this point in the history

  10. Chunk per-channel mean

    @casper-hansen
    casper-hansen committed Jun 21, 2024
    Configuration menu
    Copy the full SHA
    af92421 View commit details
    Browse the repository at this point in the history

Commits on Jun 24, 2024

  1. Merge branch 'main' into batched_quantization

    @casper-hansen
    casper-hansen committed Jun 24, 2024
    Configuration menu
    Copy the full SHA
    0565748 View commit details
    Browse the repository at this point in the history

Commits on Jun 30, 2024

  1. Optimize UX with progress bars

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    4b68287 View commit details
    Browse the repository at this point in the history

  2. Update long-context example

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    6dfdfbb View commit details
    Browse the repository at this point in the history

  3. Expose max_chunk_memory parameter

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    ca00573 View commit details
    Browse the repository at this point in the history

  4. apply code formatting

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    77401c0 View commit details
    Browse the repository at this point in the history

  5. Fix overflow: per-channel mean in FP32

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    73a96f0 View commit details
    Browse the repository at this point in the history

  6. Merge branch 'main' into batched_quantization

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    987bcfc View commit details
    Browse the repository at this point in the history

  7. Coding example

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    1f66471 View commit details
    Browse the repository at this point in the history

  8. Optimize quantization speed. Fix OOM due to inp -> inp_flat.

    @casper-hansen
    casper-hansen committed Jun 30, 2024
    Configuration menu
    Copy the full SHA
    8111285 View commit details
    Browse the repository at this point in the history

Commits on Jul 2, 2024

  1. Update coding example

    @casper-hansen
    casper-hansen committed Jul 2, 2024
    Configuration menu
    Copy the full SHA
    7ad5ac9 View commit details
    Browse the repository at this point in the history