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

    Neuronpedia

    INDEX
    Explanations

    DeepMind, Bialik, hostilis, Lewy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     adiab
    0.45
     applic
    0.44
     implic
    0.43
     élim
    0.43
     trase
    0.43
     mêmes
    0.42
     spawned
    0.42
     vivi
    0.41
     anni
    0.41
     inti
    0.41
    POSITIVE LOGITS
    illon
    0.50
    on
    0.47
    arr
    0.47
    ano
    0.46
    iles
    0.46
    art
    0.46
    ous
    0.45
    arn
    0.44
    ana
    0.44
    ford
    0.44
    Act Density 0.051%

    No Known Activations