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

    Neuronpedia

    INDEX
    Explanations

    phrases and connectors that indicate relationships or conditions in a narrative or argument

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ife
    -0.15
     etc
    -0.14
    rim
    -0.13
    uards
    -0.13
    contre
    -0.13
    ause
    -0.13
    usto
    -0.13
    addock
    -0.13
    abi
    -0.13
    rende
    -0.13
    POSITIVE LOGITS
    æk
    0.18
    /-
    0.16
    бÑĥдÑĮ
    0.15
    ifornia
    0.14
    bidden
    0.14
    ctr
    0.14
    оÑģÑĤ
    0.14
     Picker
    0.13
    cmath
    0.13
    ULA
    0.13
    Act Density 0.175%

    No Known Activations