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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     No
    0.88
     That
    0.85
     And
    0.84
    0.83
     So
    0.80
     Who
    0.77
     Do
    0.73
     As
    0.72
     Plus
    0.71
     The
    0.71
    POSITIVE LOGITS
    <unused1127>
    1.23
    <unused1063>
    1.23
    <unused204>
    1.23
    <unused1218>
    1.22
    <unused159>
    1.20
    <unused1715>
    1.20
    <unused282>
    1.19
    <unused873>
    1.19
    <unused274>
    1.18
    <unused1986>
    1.18
    Act Density 4.096%

    No Known Activations