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

    Neuronpedia

    INDEX
    Explanations

    company names

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     explicitly
    0.52
     examination
    0.50
     reluctantly
    0.50
     emptying
    0.50
     idiom
    0.49
     series
    0.49
     classification
    0.49
     height
    0.48
     ejaculation
    0.47
     accepted
    0.47
    POSITIVE LOGITS
    Easy
    0.57
    Fusion
    0.56
    Mozilla
    0.54
    Dragon
    0.53
    Photon
    0.53
    Tensor
    0.52
    Microsoft
    0.50
    Flex
    0.50
    Zen
    0.50
    Syn
    0.50
    Act Density 0.334%

    No Known Activations