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

    Neuronpedia

    INDEX
    Explanations

    phrases related to moving forward or progressing

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

    phrases indicating future intentions or actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.02
    2:0.20
    3:0.07
    4:0.04
    5:0.10
    6:0.02
    7:0.03
    8:0.11
    9:0.20
    10:0.06
    11:0.03
    Negative Logits
    essential
    -1.38
    bath
    -1.33
    afety
    -1.14
    anut
    -1.14
     Roses
    -1.14
    aneous
    -1.11
     cro
    -1.11
    organic
    -1.11
    phot
    -1.08
    ickle
    -1.07
    POSITIVE LOGITS
    ISSION
    1.21
     Zeit
    1.21
     Enhancement
    1.20
     Belt
    1.17
    ONSORED
    1.16
    ��
    1.15
    �
    1.15
     advoc
    1.11
     Kepler
    1.10
    ゴン
    1.10
    Act Density 0.026%

    No Known Activations