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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "to" in various forms, suggesting a focus on infinitive verbs or purposes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    velle
    -0.17
    umd
    -0.15
    umb
    -0.15
    pit
    -0.15
    OwnProperty
    -0.14
    otti
    -0.14
     IDC
    -0.14
    room
    -0.14
    ãĥ¼ãĤº
    -0.14
    rica
    -0.14
    POSITIVE LOGITS
    gether
    0.17
    iyan
    0.17
    bidden
    0.16
    grim
    0.15
    orny
    0.15
    hiba
    0.14
    exion
    0.14
    aken
    0.14
    alet
    0.14
    zer
    0.14
    Act Density 0.050%

    No Known Activations