© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26606
    Prev
    Next
    INDEX
    Explanations

    awards and nominations

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron detects the title-case words that name award categories (e.g. “Traditional,” “Card,” “Game,” “Graphic,” “Design”) in award listings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    mainly
    -0.99
     decidedly
    -0.98
    imee
    -0.97
    ⅼ
    -0.96
     trover
    -0.96
     insanely
    -0.95
    pherds
    -0.94
    vaila
    -0.94
     eſſ
    -0.93
    有一些
    -0.92
    POSITIVE LOGITS
     during
    1.35
     with
    1.24
     or
    1.16
     using
    1.10
     any
    1.10
     within
    1.07
    愷
    1.05
    Vanjske
    1.03
     since
    1.03
     in
    1.03
    Activations Density 0.049%

    No Known Activations