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

    Neuronpedia

    INDEX
    Explanations

    cooking states and results

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Fram
    0.73
     ബ്ല
    0.69
     Sandbox
    0.65
     വിഭാഗ
    0.64
    Mix
    0.64
    Winter
    0.63
    Blend
    0.63
    Sandbox
    0.63
     وخ
    0.62
    orthin
    0.62
    POSITIVE LOGITS
     cooking
    1.70
     cuisson
    1.60
     cook
    1.56
     frying
    1.55
     charred
    1.54
     roasting
    1.52
    cooking
    1.51
     grilling
    1.51
    cook
    1.50
     sizz
    1.48
    Act Density 0.237%

    No Known Activations