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

    Neuronpedia

    INDEX
    Explanations

    wouldn't trade it

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     taches
    0.52
    牦
    0.51
     photons
    0.49
     mussels
    0.47
     oo
    0.47
     grupos
    0.46
     ווע
    0.46
     crystals
    0.46
     particules
    0.45
     bottles
    0.45
    POSITIVE LOGITS
    price
    0.61
    rosi
    0.60
    veri
    0.55
    silver
    0.54
    しまった
    0.52
    పతి
    0.52
    state
    0.50
     ব্যবস্থ
    0.50
    contract
    0.49
    0.49
    Act Density 0.000%

    No Known Activations