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

    Neuronpedia

    INDEX
    Explanations

    numerical values preceded by symbols indicating an action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the character 'â'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    odox
    -0.68
     mushroom
    -0.66
    utenberg
    -0.61
     Osc
    -0.59
    osite
    -0.59
     bombed
    -0.57
     mushrooms
    -0.56
     Acid
    -0.56
     Horizon
    -0.56
    opolis
    -0.56
    POSITIVE LOGITS
    âĢ
    3.71
     âĢ
    2.13
    âĢł
    1.64
    âĢİ
    1.35
    âĺ
    1.34
    »
    1.33
    âľ
    1.32
    1.30
    âĹ
    1.26
    ï¸
    1.25
    Act Density 0.812%

    No Known Activations