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

Add support for BERT embedding models #5423

Merged
merged 21 commits into from
Feb 11, 2024
Merged

Add support for BERT embedding models #5423

merged 21 commits into from
Feb 11, 2024

Commits on Feb 6, 2024

  1. BERT WIP

    @cebtenzzre
    cebtenzzre committed Feb 6, 2024
    Configuration menu
    Copy the full SHA
    7286b83 View commit details
    Browse the repository at this point in the history

Commits on Feb 7, 2024

  1. merge from master

    @iamlemec
    iamlemec committed Feb 7, 2024
    Configuration menu
    Copy the full SHA
    ef10d78 View commit details
    Browse the repository at this point in the history

Commits on Feb 8, 2024

  1. it runs; tokenization is messed up; pooling is wrong for multi batches

    @iamlemec
    iamlemec committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    0051c82 View commit details
    Browse the repository at this point in the history

  2. add in wordpiece tokenizer

    @iamlemec
    iamlemec committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    59c1829 View commit details
    Browse the repository at this point in the history

  3. put causal_attn flag in gguf

    @iamlemec
    iamlemec committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    5f1c21d View commit details
    Browse the repository at this point in the history

  4. Merge remote-tracking branch 'origin/master' into bert

    @iamlemec
    iamlemec committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    e0e14e3 View commit details
    Browse the repository at this point in the history

  5. Merge remote-tracking branch 'upstream/master' into bert

    @iamlemec
    iamlemec committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    7218c7b View commit details
    Browse the repository at this point in the history

  6. Update convert-hf-to-gguf.py 

    Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
    @iamlemec @cebtenzzre
    iamlemec and cebtenzzre authored Feb 8, 2024
    Configuration menu
    Copy the full SHA
    e3efcf1 View commit details
    Browse the repository at this point in the history

  7. add causal attention gguf key

    @iamlemec
    iamlemec committed Feb 8, 2024
    Configuration menu
    Copy the full SHA
    96d37f8 View commit details
    Browse the repository at this point in the history

Commits on Feb 9, 2024

  1. use ctx_output for tok_norm of BERT and BLOOM

    @cebtenzzre
    cebtenzzre committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    e78388d View commit details
    Browse the repository at this point in the history

  2. bert : add some missing graph callbacks

    @cebtenzzre
    cebtenzzre committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    b14c457 View commit details
    Browse the repository at this point in the history

  3. fix up model sizing and result acquisition

    @iamlemec
    iamlemec committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    6875808 View commit details
    Browse the repository at this point in the history

  4. hard-code token_type = 0

    @iamlemec
    iamlemec committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    d080beb View commit details
    Browse the repository at this point in the history

  5. Merge branch 'bert' of github.com:iamlemec/llama.cpp into bert

    @iamlemec
    iamlemec committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    3a1895d View commit details
    Browse the repository at this point in the history

  6. style fixes

    @iamlemec
    iamlemec committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    961e98f View commit details
    Browse the repository at this point in the history

  7. undo attempted type_embd simplify

    @iamlemec
    iamlemec committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    56afb2f View commit details
    Browse the repository at this point in the history

  8. bert : simplify token type embedding access

    @cebtenzzre
    cebtenzzre committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    ab49e9e View commit details
    Browse the repository at this point in the history

  9. flake8 : add W503 to ignore list

    @cebtenzzre
    cebtenzzre committed Feb 9, 2024
    Configuration menu
    Copy the full SHA
    6972e7e View commit details
    Browse the repository at this point in the history

Commits on Feb 11, 2024

  1. minor : code style normalization

    @ggerganov
    ggerganov committed Feb 11, 2024
    Configuration menu
    Copy the full SHA
    8fbefed View commit details
    Browse the repository at this point in the history

  2. avoid use of ggml_graph_get_tensor

    @iamlemec
    iamlemec committed Feb 11, 2024
    Configuration menu
    Copy the full SHA
    e379e8c View commit details
    Browse the repository at this point in the history

  3. Merge branch 'bert' of github.com:iamlemec/llama.cpp into bert

    @iamlemec
    iamlemec committed Feb 11, 2024
    Configuration menu
    Copy the full SHA
    61bab47 View commit details
    Browse the repository at this point in the history