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

    Neuronpedia

    INDEX
    Explanations

    early stage and warning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Earlier
    1.13
     Earlier
    1.07
     earlier
    1.05
     tidigare
    0.99
     раньше
    0.95
    formerly
    0.95
    recent
    0.91
    Formerly
    0.90
     later
    0.90
    Later
    0.87
    POSITIVE LOGITS
     stages
    2.06
     Stages
    1.73
     stage
    1.72
    stage
    1.61
    stages
    1.61
     adopters
    1.53
     adopter
    1.50
     Stage
    1.40
    Stages
    1.30
     warning
    1.24
    Act Density 0.051%

    No Known Activations