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

    Neuronpedia

    INDEX
    Explanations

    this

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Salvation
    -0.07
    нимать
    -0.07
     furniture
    -0.07
     language
    -0.07
    ardo
    -0.07
     каждого
    -0.06
     accumulate
    -0.06
    imenti
    -0.06
    mony
    -0.06
     було
    -0.06
    POSITIVE LOGITS
    ){
↵
↵
    0.06
    -suite
    0.06
    skills
    0.06
    operands
    0.06
    	
↵
↵
    0.06
     pev
    0.06
     bif
    0.06
    국의
    0.06
     discrete
    0.06
     Ис
    0.06
    Act Density 0.009%

    No Known Activations