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

    Neuronpedia

    INDEX
    Explanations

    disjunctions and negations in the context of arguments or statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.25
    1:0.03
    2:0.03
    3:0.08
    4:0.08
    5:0.05
    6:0.07
    7:0.11
    8:0.11
    9:0.08
    10:0.01
    11:0.04
    Negative Logits
     sidelines
    -1.85
     fertile
    -1.77
     custod
    -1.65
     circus
    -1.63
     headlines
    -1.63
     Renaissance
    -1.60
     renaissance
    -1.56
     ahead
    -1.56
     spotlight
    -1.51
     fences
    -1.46
    POSITIVE LOGITS
    etheless
    2.15
     ___
    2.02
    _.
    1.93
    pg
    1.90
    netflix
    1.83
    tein
    1.80
    !,
    1.76
     Expect
    1.75
    say
    1.73
    fml
    1.72
    Act Density 0.002%

    No Known Activations