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

    Neuronpedia

    INDEX
    Explanations

    interests and personality

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     formulations
    0.54
     formalism
    0.52
     consequences
    0.47
     definitions
    0.47
     procedures
    0.46
     algorithms
    0.46
     umož
    0.46
     reap
    0.46
     obfusc
    0.46
     panels
    0.46
    POSITIVE LOGITS
     hobbies
    0.83
     любит
    0.82
     Personality
    0.75
    性格
    0.75
     좋아하는
    0.75
    Hobbies
    0.74
     Interests
    0.74
     люблю
    0.71
     loves
    0.71
    喜欢
    0.68
    Act Density 0.141%

    No Known Activations