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

    Neuronpedia

    INDEX
    Explanations

    conditions and actions that are contingent on specific circumstances

    oai_token-act-pair · gpt-3.5-turbo

    conditional phrases or statements prefaced by "unless" followed by actions or situations that the subject may find themselves in

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĥ´
    -0.71
    word
    -0.68
    VG
    -0.66
    ãĥ³
    -0.62
    terson
    -0.60
    Topics
    -0.60
    same
    -0.59
    ãĥ«
    -0.59
     unanswered
    -0.58
    no
    -0.58
    POSITIVE LOGITS
     expressly
    1.01
     explicitly
    0.94
     specifically
    0.85
     somehow
    0.84
     absolutely
    0.81
    pecially
    0.77
     otherwise
    0.74
     interven
    0.73
     urgently
    0.72
     willfully
    0.72
    Act Density 0.151%

    No Known Activations