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

    Neuronpedia

    INDEX
    Explanations

    error handling and debugging

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     devotees
    1.06
     burgeoning
    1.03
     now
    1.00
     juncture
    0.94
     provocation
    0.94
     temptations
    0.94
     aftermath
    0.93
     porch
    0.93
     conviction
    0.90
     predicament
    0.90
    POSITIVE LOGITS
    Error
    0.98
    没有
    0.92
    Delete
    0.92
    Remove
    0.90
    Copy
    0.90
    图片
    0.89
    Debug
    0.89
    Method
    0.87
    Update
    0.86
    debug
    0.84
    Act Density 0.057%

    No Known Activations