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

    Neuronpedia

    INDEX
    Explanations

    this

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Few
    0.33
    几
    0.32
     Validity
    0.32
     Idee
    0.32
     Requirement
    0.31
     Plusieurs
    0.31
     числі
    0.31
    ༌
    0.31
     Einige
    0.30
    幾個
    0.29
    POSITIVE LOGITS
     particular
    0.85
    particular
    0.64
     particolare
    0.63
     unassuming
    0.59
     represents
    0.55
     intriguing
    0.54
     ensures
    0.53
     relatively
    0.53
     incredible
    0.51
     particulière
    0.50
    Act Density 0.432%

    No Known Activations