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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences structured with "âĢĶ" or similar special characters

    oai_token-act-pair · gpt-3.5-turbo

    different forms of an unusual symbol or character used in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤ©
    -0.71
     guiActiveUnfocused
    -0.61
    enta
    -0.59
    ãĥīãĥ©
    -0.59
     Supported
    -0.57
     )))
    -0.57
    utive
    -0.56
    èĢħ
    -0.56
     0004
    -0.54
    peg
    -0.54
    POSITIVE LOGITS
    especially
    1.26
    particularly
    1.14
     namely
    1.12
    albeit
    1.07
     albeit
    1.06
    until
    1.06
    although
    1.04
    except
    1.02
    unless
    1.01
    they
    1.01
    Act Density 0.148%

    No Known Activations