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

    Neuronpedia

    INDEX
    Explanations

    contractions of "do not" followed by a verb

    oai_token-act-pair · gpt-3.5-turbo

    negations and expressions of reluctance or refusal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     populated
    -0.68
     Alas
    -0.66
     eleph
    -0.66
     Adv
    -0.66
     princ
    -0.65
     nearest
    -0.60
     eagerly
    -0.59
     aims
    -0.59
    anwhile
    -0.58
     HF
    -0.58
    POSITIVE LOGITS
    't
    1.64
    ÃŃ
    0.96
    nis
    0.88
    uts
    0.84
    etsk
    0.84
    ovan
    0.82
    iting
    0.82
    nat
    0.81
    ´
    0.80
    n
    0.79
    Act Density 0.116%

    No Known Activations