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

    Neuronpedia

    INDEX
    Explanations

    eliminate or remove

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     slowing
    0.47
     slowed
    0.46
     slowdown
    0.46
     diversion
    0.44
     ralent
    0.41
     prolongation
    0.40
     postponement
    0.39
    Failed
    0.38
     আহত
    0.38
     deceleration
    0.38
    POSITIVE LOGITS
     eliminate
    2.86
    Elim
    2.70
     eliminated
    2.67
     eliminating
    2.67
     eliminates
    2.67
     elimin
    2.66
     Eliminate
    2.64
     elimination
    2.61
    eliminate
    2.59
     élim
    2.56
    Act Density 0.137%

    No Known Activations