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

    Neuronpedia

    INDEX
    Explanations

    degree or level of abstract quality

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     workflows
    1.20
     extremamente
    1.11
     самые
    1.10
     extremadamente
    1.06
     primeiras
    1.05
     altamente
    1.04
     большие
    1.04
     малень
    1.03
     soluciones
    1.02
     மிகப்பெரிய
    1.02
    POSITIVE LOGITS
     interest
    1.15
     difficulty
    1.09
     concern
    1.08
     excitement
    1.07
     apprehension
    1.02
     interesse
    0.99
    interest
    0.97
     enthusiasm
    0.97
     suspicion
    0.96
     credence
    0.95
    Act Density 0.259%

    No Known Activations