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

    Neuronpedia

    INDEX
    Explanations

    expressions involving expectations or hypothetical situations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Sandwich
    -0.60
    ammy
    -0.60
     revived
    -0.59
     demolished
    -0.58
    stood
    -0.57
     Cance
    -0.57
    sei
    -0.56
     reopened
    -0.55
    usha
    -0.55
    isen
    -0.55
    POSITIVE LOGITS
     expect
    1.17
     normally
    1.15
     ordinarily
    1.09
     otherwise
    1.08
     imagine
    0.92
     characterize
    0.92
     anticipate
    0.85
     tolerate
    0.85
     deem
    0.84
     prefer
    0.84
    Act Density 0.104%

    No Known Activations