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

    Neuronpedia

    INDEX
    Explanations

    arrows and function definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     after
    0.98
     through
    0.96
     than
    0.91
    through
    0.82
     more
    0.79
    after
    0.78
     final
    0.78
    than
    0.75
     niches
    0.74
     finally
    0.72
    POSITIVE LOGITS
     =>
    1.87
    =>
    1.56
     =&
    1.32
     =>{
    1.29
    Rightarrow
    1.19
    =>$
    1.13
     ->
    1.12
     ==>
    1.12
    Retour
    1.10
    =>{
    1.09
    Act Density 0.029%

    No Known Activations