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

    Neuronpedia

    INDEX
    Explanations

    words related to actions or concepts involving strong emotions or impactful decisions

    oai_token-act-pair · gpt-3.5-turbo

    actions and processes related to acceptance and management

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    usually
    -0.66
     cutter
    -0.63
     generally
    -0.61
     equally
    -0.60
    chief
    -0.59
     chiefly
    -0.58
     basin
    -0.58
     obviously
    -0.58
    again
    -0.58
     respectively
    -0.57
    POSITIVE LOGITS
    Initialized
    0.85
    ãĥīãĥ©
    0.81
    ounters
    0.74
    ifiable
    0.74
    arded
    0.72
    enance
    0.68
    rencies
    0.67
    ÃŁ
    0.66
     fault
    0.66
    kson
    0.65
    Act Density 0.518%

    No Known Activations