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

    Neuronpedia

    INDEX
    Explanations

    evaluative adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Charcoal
    0.52
     Modular
    0.50
     Rustic
    0.49
     restructured
    0.49
     Starter
    0.48
     Macrom
    0.48
     Multi
    0.47
     Combo
    0.46
     Handles
    0.46
     Mult
    0.45
    POSITIVE LOGITS
     horrifying
    0.58
     irony
    0.54
     circumstances
    0.53
     sickening
    0.52
     disbelief
    0.52
    erweise
    0.52
     heartbreaking
    0.50
     ironic
    0.50
     hypocrisy
    0.49
    竟然
    0.49
    Act Density 0.094%

    No Known Activations