Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. GPT2-Small
    3. Transcoders Residuals
    4. 9-TRES-DC
    5. 613
    Prev
    Next
    INDEX
    Explanations

    instances of the prefix "un-" indicating negation or absence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     drones
    -0.73
     sail
    -0.71
     submarine
    -0.70
     drone
    -0.70
     sails
    -0.67
     echoing
    -0.66
     trumpet
    -0.66
     disconnected
    -0.65
     interchangeable
    -0.65
     distracting
    -0.64
    POSITIVE LOGITS
    season
    0.94
    god
    0.92
    recogn
    0.90
    prov
    0.85
    aware
    0.84
    character
    0.84
    anth
    0.83
    gu
    0.83
    fort
    0.82
    har
    0.81
    Activations Density 0.023%

    No Known Activations