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

    Neuronpedia

    INDEX
    Explanations

    max_

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     legions
    0.41
     underpinned
    0.40
     fussy
    0.40
     purposeful
    0.39
     ingrained
    0.39
     seep
    0.39
     overly
    0.39
     walled
    0.39
     cliché
    0.39
     heresy
    0.39
    POSITIVE LOGITS
    <unused60>
    0.48
    _
    0.42
    :=
    0.37
    <unused63>
    0.36
    <0x1D>
    0.35
    ]));
    0.35
    
    0.34
    <0x1C>
    0.34
    _{
    0.34
    "/>
    0.33
    Act Density 6.621%

    No Known Activations