© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20059
    Prev
    Next
    INDEX
    Explanations

    act of forgetting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     zapal
    -2.06
    漚
    -1.98
    
    -1.91
    鞮
    -1.84
     články
    -1.82
    íte
    -1.82
     ander
    -1.82
    
    -1.80
     THOSE
    -1.80
     THEIR
    -1.73
    POSITIVE LOGITS
     forgot
    2.17
     forget
    2.08
    or
    2.02
     that
    1.95
     forgets
    1.90
    .
    1.84
     Forget
    1.84
     what
    1.83
     threatening
    1.80
    in
    1.77
    Activations Density 0.015%

    No Known Activations