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

    Neuronpedia

    INDEX
    Explanations

    introduces advice or a question

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Addressing
    0.69
    prehensive
    0.61
    ilibus
    0.61
    Exploring
    0.60
     преимущественно
    0.58
    extensive
    0.57
     belirli
    0.57
    acağız
    0.57
    uzioni
    0.56
    Copies
    0.56
    POSITIVE LOGITS
     statement
    2.05
     comment
    1.68
     sentence
    1.66
     message
    1.64
    statement
    1.56
     anecdote
    1.51
     assertion
    1.50
     Statement
    1.49
     idea
    1.48
     scenario
    1.48
    Act Density 2.404%

    No Known Activations