Commit History

  1. Add code revision argument and validate launch inputs for HF job script⁠

    Mohith202 +1: Copilot
    1e93c48
  2. Merge branch 'SI_31' into main

    285d621
  3. Refactor HF job script: remove unused functions, update token handling, and enhance argument parser for clarity⁠

    Mohith202 +1: Copilot
    6f7c943
  4. Add functions for handling Hugging Face token and default repository URLs; update argument parser defaults⁠

    Mohith202 +1: Copilot
    c1872ee
  5. Enhance job script logging and environment setup for improved clarity during execution

    e66e1d8
  6. Refactor code structure for improved readability and maintainability and also added the mask ROI⁠

    Mohith202 +1: Copilot
    0dac2bf

  1. Refactor A1 evaluation protocols to focus exclusively on Protocol C; update documentation and scripts accordingly.

    e5d02d5
  2. Skip LoRA SFT for LLaDA (masked diffusion LM, incompatible with causal SFT); pass through HF ID

    17ccdd5
  3. Fix TRL compat: pre-tokenize dataset, drop max_seq_length from SFTConfig/SFTTrainer

    42ac627
  4. Fix SFTConfig/SFTTrainer API compat across TRL versions (max_seq_length, processing_class)

    ae9a13f
  5. Fix duplicate --feature-num-workers/--num-fit-workers args in parser

    d3b1138
  6. Enhance feature extraction with multi-GPU support and worker resolution; add parallel processing for fitting layers

    4f2bff8
  7. Add instruct-tuning (LoRA), multi-model fit, comparison viz; fix missing flags

    279c017

  1. added the test run for HF job

    3f888c6
  2. Add core ROI masks, visualization script, and model profiles⁠

    8c5a642
  3. Merge hf/master: restore run_a1 files and hf_jobs

    33c7212
  4. initial commit

    Verified signature
    485cc1e