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

    Neuronpedia

    INDEX
    Explanations

    phrases related to legal and medical topics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    various forms of pronunciations and statements related to moral or ethical assessments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.04
    2:0.17
    3:0.05
    4:0.18
    5:0.07
    6:0.03
    7:0.03
    8:0.11
    9:0.11
    10:0.06
    11:0.03
    Negative Logits
    adier
    -1.23
    EStreamFrame
    -1.15
    aughty
    -1.09
    oine
    -1.08
     edges
    -1.06
     Reloaded
    -1.04
     lookout
    -1.03
     payroll
    -1.03
    FTWARE
    -1.02
    ruce
    -1.01
    POSITIVE LOGITS
     aloud
    1.31
    aults
    1.25
    ript
    1.23
     uttered
    1.15
     ital
    1.15
     ut
    1.15
     disapp
    1.15
     paras
    1.10
     pronoun
    1.09
     Scots
    1.09
    Act Density 0.003%

    No Known Activations