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

    Neuronpedia

    INDEX
    Explanations

    fuel efficiency ratings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ছিল
    0.43
    ('/\
    0.38
    Opts
    0.36
     eher
    0.35
     열심히
    0.35
    }]=
    0.34
    惧
    0.34
    Sampler
    0.34
     пое
    0.34
    western
    0.34
    POSITIVE LOGITS
     get
    0.40
     bil
    0.37
     Get
    0.37
     default
    0.36
     स्म
    0.36
     rating
    0.36
    rab
    0.35
     health
    0.35
     thirty
    0.35
     Egyes
    0.35
    Act Density 0.001%

    No Known Activations