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

    Neuronpedia

    INDEX
    Explanations

    diffusion models denoising

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rohkem
    0.45
     enforcing
    0.44
     bling
    0.40
     murderous
    0.39
     educating
    0.39
     instructive
    0.39
     autres
    0.39
     destitute
    0.39
     subst
    0.39
     prots
    0.39
    POSITIVE LOGITS
    alf
    0.42
     भएका
    0.41
    onat
    0.39
     טי
    0.38
    ona
    0.38
    bev
    0.38
    ion
    0.37
    entuan
    0.37
    anoj
    0.36
    one
    0.36
    Act Density 0.001%

    No Known Activations