© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. GPT2-Small
    5. Transcoders Residuals
    7. 8-TRES-DC
    9. 3335
    Prev
    Next
    INDEX
    Explanations

    negative contractions related to denial or refusal

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Negative Logits
     inaccur
    -0.79
     negatives
    -0.72
     confounding
    -0.72
     stigma
    -0.71
     paradox
    -0.71
     unfairly
    -0.70
     offend
    -0.69
     worries
    -0.68
     unwanted
    -0.67
     worrying
    -0.66
    POSITIVE LOGITS
    weekly
    0.68
    atu
    0.68
    vati
    0.65
     Ready
    0.64
    shi
    0.64
    geon
    0.63
     Vaughn
    0.63
    gamer
    0.63
     Theodore
    0.63
    complete
    0.62
    Activations Density 0.228%

    No Known Activations