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

    Neuronpedia

    INDEX
    Explanations

    verbs associated with planned or expected actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating actions and expectations regarding future events or statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abiding
    -0.90
     maximal
    -0.69
     Endless
    -0.69
    selves
    -0.68
     Definitions
    -0.68
    /-
    -0.66
     genesis
    -0.65
    humans
    -0.64
     definitions
    -0.63
     Decay
    -0.63
    POSITIVE LOGITS
     resign
    1.14
     personally
    1.14
     testify
    1.11
     consult
    1.11
     attend
    1.11
     confess
    1.07
     plead
    1.07
     announce
    1.06
     marry
    1.06
     verbally
    1.05
    Act Density 0.383%

    No Known Activations