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

    Neuronpedia

    INDEX
    Explanations

    fit well or best

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    对
    0.60
     trust
    0.59
     reach
    0.59
     upon
    0.58
     contact
    0.55
     महापौर
    0.55
     let
    0.55
    తూ
    0.54
     contacts
    0.52
    对其
    0.51
    POSITIVE LOGITS
     snugly
    1.08
     mieux
    1.03
     잘
    1.03
     отлично
    1.02
     dobře
    1.02
     ótimo
    1.01
     better
    1.00
     хорошо
    0.99
     melhor
    0.99
     beter
    0.97
    Act Density 0.120%

    No Known Activations