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

    Neuronpedia

    INDEX
    Explanations

    encoding input

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     beam
    0.54
    beam
    0.51
     generator
    0.49
     Beam
    0.49
    Beam
    0.49
     generation
    0.48
     beams
    0.48
    Generation
    0.46
     génération
    0.45
     generazione
    0.43
    POSITIVE LOGITS
     Encoding
    0.64
     encoding
    0.58
     Encode
    0.57
    Encode
    0.56
     encode
    0.55
     encoded
    0.52
    编码
    0.50
    encoding
    0.50
     encodes
    0.48
    Encoding
    0.47
    Act Density 0.005%

    No Known Activations