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

    Neuronpedia

    INDEX
    Explanations

    conditional phrases, "if" and "or"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ]
    0.39
    swith
    0.37
    ్య
    0.36
    }
    0.36
    sof
    0.35
    Garden
    0.35
    working
    0.35
    Si
    0.35
    ことから
    0.34
     करीब
    0.34
    POSITIVE LOGITS
     hãy
    0.50
     અથવા
    0.47
     किंवा
    0.45
     chances
    0.43
     etc
    0.43
     وغیرہ
    0.43
     அல்லது
    0.43
    那就
    0.42
     alebo
    0.42
     shouldn
    0.41
    Act Density 0.399%

    No Known Activations