TY - DATA
T1 -  Software and data underlying the publication: "Position: Stop Making Unscientific AGI Performance Claims"
PY - 2025/07/15
AU - Patrick Altmeyer
AU - C.C.S. (Cynthia) Liem
AU - Andrew Demetriou
AU - Antony Bartlett
UR - 
DO - 10.4121/d427d182-4bb0-4972-980c-adcb28f430b6.v1
KW - Artificial Intelligence
KW - Artificial General Intelligence
KW - Mechanistic Interpretability
KW - Interpretability
N2 - <p>Code and research results for ICML 2024 position paper. Originally released here: https://github.com/pat-alt/spurious_sentience.</p><p><br></p><p>The research results include:</p><p><br></p><ol><li>Regression tables (.tex; .html)</li><li>An "evaluations.csv" file that contains estimated evaluation metrics for linear probes  and the baseline grouped by indicator, layer (network layer), train/test split, variable (measure), model (lin. probe/baseline).</li><li>A figures/ folder containing all PNG figures that went into a) the body or b) the appendix.</li><li>An interim/ folder containing results for probe predictions for each training epoch.</li><li>An attacks/ folder containing the CSV files of neural network activations for attack prompts (see paper for details). Additionally, this folder contains a sentences/ subfolder with the actual textual attack prompts (.txt files).</li></ol><p><br></p>
ER -