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

    Neuronpedia

    INDEX
    Explanations

    enhanced + technical terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     chips
    0.42
     inputs
    0.41
     incoming
    0.40
     ou
    0.37
     crossed
    0.37
     arrivals
    0.37
     shone
    0.37
     efforts
    0.37
     chip
    0.37
    となって
    0.37
    POSITIVE LOGITS
     Instructor
    0.54
    Instructor
    0.53
     Transcription
    0.51
    RESOLUTION
    0.45
    Instruct
    0.44
    instructor
    0.43
    ല്ല
    0.43
    usian
    0.42
    اداسی
    0.42
    鴳
    0.42
    Act Density 0.001%

    No Known Activations