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

    Neuronpedia

    INDEX
    Explanations

    phrases related to evaluation and assessment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    preceding "is" or "be"

    np_acts-logits-general · gemini-2.0-flash

    selling technique

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tryna
    -1.09
    Portail
    -1.05
    Alongside
    -1.02
     fucked
    -1.00
     shitty
    -1.00
     prioritise
    -0.99
     :')
    -0.97
     weirdly
    -0.96
    idk
    -0.94
     jspb
    -0.94
    POSITIVE LOGITS
     muß
    0.78
     idéia
    0.72
     skall
    0.71
     läßt
    0.65
    spania
    0.65
    ㄋ
    0.63
     monies
    0.63
    0.61
     Moslem
    0.61
    ・・・・・
    0.60
    Act Density 1.745%

    No Known Activations