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

    Neuronpedia

    INDEX
    Explanations

    tunnel

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    1.34
    ни
    1.01
    ak
    1.00
    k
    0.97
    2
    0.88
    ’
    0.84
    ką
    0.82
    ের
    0.82
    да
    0.81
    p
    0.81
    POSITIVE LOGITS
     tunnels
    1.66
     tunnel
    1.47
     Tunnel
    1.37
    Tunnel
    1.28
     tunneling
    1.25
    隧道
    1.15
     tunnelling
    1.12
     underground
    0.93
     burrow
    0.93
     tunn
    0.91
    Act Density 0.006%

    No Known Activations