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

    Neuronpedia

    INDEX
    Explanations

    contains all of the following

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    zyna
    1.33
    chid
    1.24
    diendo
    1.20
    dyž
    1.19
     fired
    1.15
    udas
    1.15
    avond
    1.13
    houses
    1.13
    よう
    1.12
     chitosan
    1.11
    POSITIVE LOGITS
    imately
    1.59
     RSM
    1.44
     والح
    1.35
    ৩
    1.32
     Rocky
    1.28
     snapshots
    1.28
    시키는
    1.27
    ahr
    1.27
     zod
    1.27
    𝘻
    1.26
    Act Density 0.000%

    No Known Activations