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

    Neuronpedia

    INDEX
    Explanations

    Lubenian

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     loosing
    0.50
     (~
    0.42
     Beside
    0.39
    ـ
    0.39
    Notice
    0.38
    And
    0.37
     –
    0.37
     planing
    0.37
    ometime
    0.36
     McConnell
    0.35
    POSITIVE LOGITS
     Pursuant
    0.44
    🪐
    0.44
     коронави
    0.44
     великолеп
    0.44
     nonnegative
    0.42
     ascertaining
    0.41
     ஏராளமான
    0.41
     subpar
    0.40
     magní
    0.40
    🩷
    0.39
    Act Density 0.003%

    No Known Activations