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

    Neuronpedia

    INDEX
    Explanations

    conditional statements or phrases indicating situations that depend on specific criteria

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     if
    -0.15
     behalf
    -0.15
    ilden
    -0.14
     basically
    -0.14
    ky
    -0.14
     wenn
    -0.14
    er
    -0.13
     unless
    -0.13
    常
    -0.13
    occo
    -0.13
    POSITIVE LOGITS
     possible
    0.26
     necessary
    0.24
     Necessary
    0.23
    necessary
    0.22
     unsure
    0.22
    Possible
    0.21
    possible
    0.21
     posible
    0.21
     Possible
    0.20
    _possible
    0.20
    Act Density 0.109%

    No Known Activations