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

    Neuronpedia

    INDEX
    Explanations

    phrases that begin with "some" to indicate unspecified quantities or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    udden
    -0.15
    ibold
    -0.14
    oku
    -0.14
    erece
    -0.14
    asing
    -0.13
    ÑĤаж
    -0.13
    awai
    -0.13
    anders
    -0.13
    °ëĭ¤
    -0.13
    .psi
    -0.13
    POSITIVE LOGITS
     might
    0.35
     may
    0.29
     would
    0.27
     people
    0.26
     say
    0.25
     are
    0.23
     will
    0.22
    人ãģ¯
    0.21
     readers
    0.21
     times
    0.20
    Act Density 0.103%

    No Known Activations