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

    Neuronpedia

    INDEX
    Explanations

    deliberate use or action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     p
    0.90
     jellem
    0.89
    jším
    0.87
    क्रम
    0.83
     Durch
    0.81
     чем
    0.80
     greet
    0.79
    しまい
    0.79
     automatic
    0.78
     accused
    0.78
    POSITIVE LOGITS
    ം
    1.23
     roughness
    1.14
     provocative
    1.11
    Created
    1.11
     omission
    1.10
    रित
    1.09
     provocation
    1.09
    symmetry
    1.06
    د
    1.06
    ar
    1.04
    Act Density 0.101%

    No Known Activations