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

    Neuronpedia

    INDEX
    Explanations

    speeding up processes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Safety
    0.57
    es
    0.56
     Integrity
    0.53
     integrity
    0.52
     efficacy
    0.52
    is
    0.52
    yl
    0.52
     Statistical
    0.51
     Advocate
    0.51
    um
    0.50
    POSITIVE LOGITS
     aceler
    0.78
     روند
    0.75
     hastened
    0.73
     hasten
    0.70
     accél
    0.70
     уско
    0.69
     acceler
    0.65
     accelerate
    0.64
     accelerated
    0.64
     ускори
    0.63
    Act Density 0.128%

    No Known Activations