Skip to content
View hc495's full-sized avatar
💤
Sleepy Everyday
💤
Sleepy Everyday

Highlights

  • Pro

Block or report hc495

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
hc495/README.md

Working on NN/LLM Interpretability in RIKEN. Refer to https://www.hakaze-c.com/.

List of Official Code Implementation of Original Papers

  1. [ICLR 2026] Mechanism of Task-oriented Information Removal in In-context Learning
    Paper Repository
  2. Binary Autoencoder for Mechanistic Interpretability of Large Language Models
    Paper Repository
  3. [ICLR 2025] Revisiting In-context Learning Inference Circuit in Large Language Models
    Paper Repository
  4. [NAACL 2025 Main] Token-based Decision Criteria Are Suboptimal in In-context Learning
    Paper Repository
  5. [EMNLP 2025 BlackBox NLP workshop] Mechanistic Fine-tuning for In-context Learning
    Paper Repository
  6. [COLING 2025] Understanding Token Probability Encoding in Output Embeddings
    Paper Repository
  7. StaICC: Standardized Toolkit for In-context Classification
    Paper Repository PyPl
  8. NoisyICL: A Little Noise in Model Parameters Calibrates In-context Learning
    Paper Repository

Pinned Loading

  1. Verb_subspace Verb_subspace Public

    [ICLR 2026] Official code implementation of paper: "Mechanism of Task-oriented Information Removal in In-context Learning"

    Jupyter Notebook 6

  2. ICL_Circuit ICL_Circuit Public

    [ICLR 2025] Official code implementation of paper: "Revisiting In-context Learning Inference Circuit in Large Language Models"

    Jupyter Notebook 4

  3. Hidden_Calibration Hidden_Calibration Public

    [NAACL 2025 main] Official code implementation of paper: "Token-based Decision Criteria Are Suboptimal in In-context Learning"

    Jupyter Notebook 3

  4. ICL_head_tuning ICL_head_tuning Public

    [EMNLP 2025 BlackBox NLP workshop] Official code implementation of paper: "Mechanistic Fine-tuning for In-context Learning"

    Python

  5. Binary_Autoencoder Binary_Autoencoder Public

    Official code implementation of paper: "Binary Autoencoder for Mechanistic Interpretability of Large Language Models"

    Jupyter Notebook

  6. StaICC StaICC Public

    A standardized toolkit for classification task on In-context Learning. Official code implementation of paper: "StaICC: Standardized Evaluation for Classification Task in In-context Learning"

    Python 2