메뉴
Activates when working on mechanistic interpretability of transformers, specifically for automated circuit discovery in transformer models using computational graph editing and analysis.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Representation and Parameter Analysis interprets and controls a model by directly manipulating its two kinds of internal objects — features (the hidden-state activations produced during a forward pass) and weights (the parameters of the target model). The analysis side relies on standard linear-algebra tools applied straight to these objects: techniques such as PCA, mean-difference, or simple linear classifiers are run over collected features (e.g., from contrastive prompt pairs) or over weight differences between a fine-tuned and a pre-trained checkpoint, in order to extract a small set of meaningful directions. The control side then operates directly through arithmetic on those objects: adding such a direction back into the features steers behavior at inference time, while adding, subtracting, or combining task-level weight differences edits the model in one shot — so that interpretability findings translate into behavioral control without any retraining.
Use this skill for feature-level steering of models — locating the internal feature that drives a target behavior, scoring and selecting it by its effect on the model's output, and directly amplifying or shrinking that feature's activation during generation to control behavior. Applies to features read from the model's own activations or from a Sparse Autoencoder (SAE); the bundled demo scripts happen to use an SAE, but the method does not require one.
Routing entry point for eleven families of mechanistic-interpretability methods that localize *which* internal object (layer, attention head, neuron, SAE feature, weight, or input feature) drives a model's behavior, *how influential* it is, and *what changes* when it is intervened on. Use this skill whenever the question is about a model's internal mechanism rather than its external metrics — for example, claims that a specific component is responsible for a behavior, mechanistic-evidence requests, circuit-discovery tasks, feature-attribution work (SHAP), or concept-level explanations of vision/VL models. The file lays out each family's premise, signal, cost, advantages, limitations, and how to compose them into a cheap-screen → causal-verify pipeline. Loading is hierarchical and mandatory: after picking a family from this routing file you MUST load that family's `SKILL.md`, and after picking a submethod you MUST load that submethod's `SKILL.md` — never act on the previews in this file alone.
Vocabulary Projection methods interpret internal model states by projecting them through the unembedding matrix to obtain a distribution over the vocabulary. The core idea is that the unembedding matrix, which maps the final hidden state to output logits, can serve as a universal decoder for intermediate states throughout the model. By inspecting the top-ranked tokens of the resulting distribution, researchers can directly read off the semantic content encoded in any internal object — such as a residual stream state or an attention head output — in terms of the model's output vocabulary.
Use when main results pass result-to-claim (claim_supported=yes or partial) and ablation studies are needed for paper submission. The external LLM reviewer (via llm-chat MCP) designs ablations from a reviewer's perspective, CC reviews feasibility and implements.
Search, download, and summarize academic papers from arXiv. Use when user says "search arxiv", "download paper", "fetch arxiv", "arxiv search", "get paper pdf", or wants to find and save papers from arXiv to the local paper library.
| name | automatic-circuit-discovery |
| description | Activates when working on mechanistic interpretability of transformers, specifically for automated circuit discovery in transformer models using computational graph editing and analysis. |
scripts/acdc_run_demo.py#!/usr/bin/env python3
"""
Simple runnable script demonstrating how to run the main ACDC pipeline programmatically.
This script shows how to import and execute the main function in the ACDC library,
which runs the automated circuit discovery for a default configuration.
Requires:
- Python 3.8+ environment with the Automatic-Circuit-Discovery repo installed via Poetry
- System dependencies (graphviz, etc.) installed per instructions
Usage:
python scripts/acdc_run_demo.py
"""
import sys
import os
import argparse
from acdc import main as acdc_main_module
def main():
"""
Runs the main ACDC experiment pipeline, simulating the command line interface call.
Prints progress and handles basic errors.
"""
try:
# The main.py in acdc offers a CLI main function, here we call directly
# It can take command-line like arguments, but defaults should run a demo
# For example: python acdc/main.py --help for CLI instructions
# We run it with no additional args to start the default pipeline/demo
print("Starting ACDC main pipeline demo run...")
sys.argv = ['acdc/main.py'] # Reset argv for main.py
acdc_main_module.main()
print("ACDC main pipeline finished successfully.")
except Exception as e:
print(f"Error running ACDC main pipeline: {e}")
sys.exit(1)
if __name__ == "__main__":
main()