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

    Neuronpedia

    INDEX
    Explanations

    phrases related to determining or evaluating something

    oai_token-act-pair · gpt-3.5-turbo

    repeated references to "the" as a common structural element across different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    âĿ
    -0.69
    note
    -0.68
    frog
    -0.67
    tumblr
    -0.66
    Save
    -0.66
    hari
    -0.66
    NB
    -0.65
    OTS
    -0.65
    SPONSORED
    -0.63
    quished
    -0.63
    POSITIVE LOGITS
     extent
    1.59
     effectiveness
    1.37
     amount
    1.35
     likelihood
    1.34
     severity
    1.32
     outcome
    1.28
     adequ
    1.25
     direction
    1.25
     availability
    1.23
     efficacy
    1.21
    Act Density 0.390%

    No Known Activations