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

    Neuronpedia

    INDEX
    Explanations

    researchers' discoveries and studies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     becoming
    0.43
    Becoming
    0.43
    成为
    0.42
     Being
    0.40
     Drinks
    0.39
     Becoming
    0.39
    作为
    0.38
    Does
    0.38
     olur
    0.38
     Myths
    0.38
    POSITIVE LOGITS
     specializing
    0.55
     studying
    0.55
    including
    0.49
     overseeing
    0.49
     encarg
    0.49
     specialising
    0.47
     estudar
    0.46
     odpowiedzial
    0.46
     tasked
    0.45
     analyzing
    0.45
    Act Density 0.014%

    No Known Activations