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

    Neuronpedia

    INDEX
    Explanations

    occurrences of specific terms related to legal or criminal actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.11
    1:0.07
    2:0.07
    3:0.08
    4:0.04
    5:0.11
    6:0.07
    7:0.03
    8:0.07
    9:0.14
    10:0.09
    11:0.07
    Negative Logits
     scrut
    -1.67
    htaking
    -1.47
     sustainability
    -1.43
     behavi
    -1.33
     incent
    -1.27
     mitigation
    -1.26
     lifes
    -1.25
     newsp
    -1.24
     welf
    -1.23
     lapt
    -1.22
    POSITIVE LOGITS
    /,
    1.63
    -.
    1.51
    /.
    1.49
    ([
    1.41
    �
    1.40
    .[
    1.35
    /?
    1.34
    �
    1.34
    /"
    1.33
    �
    1.32
    Act Density 0.029%

    No Known Activations