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

    Neuronpedia

    INDEX
    Explanations

    expressing disappointment

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Mini
    -0.07
     suicide
    -0.07
    Accent
    -0.06
    Frozen
    -0.06
     sû
    -0.06
    Fold
    -0.06
    練
    -0.06
     dan
    -0.06
     zich
    -0.06
     dipped
    -0.06
    POSITIVE LOGITS
     multiply
    0.08
    لك
    0.08
    τές
    0.07
    (models
    0.06
    abyte
    0.06
     /\.
    0.06
     Commands
    0.06
    ีส
    0.06
    の上
    0.06
     TAS
    0.06
    Act Density 0.088%

    No Known Activations