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

    Neuronpedia

    INDEX
    Explanations

    LLaMA and quantization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Bread
    0.77
     Bread
    0.76
     Dough
    0.69
    ディース
    0.69
     AdWords
    0.69
     bread
    0.69
    Typography
    0.68
     Designers
    0.68
    上线
    0.67
     শিল
    0.67
    POSITIVE LOGITS
    llama
    1.18
     llama
    1.15
    quant
    1.02
     Llam
    1.00
     quant
    1.00
     Ll
    0.97
     quantized
    0.95
    Quant
    0.92
     cuant
    0.91
     L
    0.91
    Act Density 0.838%

    No Known Activations