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

    Neuronpedia

    INDEX
    Explanations

    inject scripts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ibh
    0.84
     Bui
    0.75
     Nieto
    0.75
    継
    0.75
     benchmarks
    0.74
    Buildable
    0.73
     hierarchy
    0.71
     bezier
    0.71
     benchmark
    0.70
    tuk
    0.69
    POSITIVE LOGITS
     script
    1.55
    Script
    1.53
    script
    1.52
     scripts
    1.46
     Script
    1.45
    scripts
    1.32
     Scripts
    1.31
     scripting
    1.29
    脚本
    1.25
    SCRIPT
    1.19
    Act Density 0.115%

    No Known Activations