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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 54354
    Prev
    Next
    INDEX
    Explanations

    United States Court of Appeals

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

    The neuron strongly activates on long runs of whitespace or indentation characters used as formatting padding at the beginnings of lines (blank‐like indent blocks).

    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
     region
    -1.10
     its
    -1.10
    老闆
    -1.08
    Ώ
    -1.05
     lengte
    -1.02
     komfor
    -1.01
     certificación
    -1.00
     suplementos
    -0.98
     大人
    -0.97
     eerste
    -0.97
    POSITIVE LOGITS
     ,‬
    1.20
     армии
    1.09
     fédé
    1.08
     statunit
    1.05
    uelles
    1.03
    Bety
    1.03
    Also
    1.02
     vinto
    1.02
     Outro
    1.01
    这
    0.99
    Activations Density 0.030%

    No Known Activations