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

    Neuronpedia

    INDEX
    Explanations

    phrases related to being forced to do something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate being forced to take specific actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    icious
    -0.76
    ergy
    -0.71
    /-
    -0.69
     Blueprint
    -0.67
    productive
    -0.65
    mal
    -0.64
    Recommended
    -0.64
     Beam
    -0.64
     Signal
    -0.63
     Done
    -0.63
    POSITIVE LOGITS
     confront
    1.17
     endure
    1.13
     rethink
    1.07
     reconsider
    1.02
     reckon
    1.02
     undergo
    0.99
     apologise
    0.94
     swallow
    0.93
     retire
    0.92
     cancel
    0.92
    Act Density 0.090%

    No Known Activations