© 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. 25752
    Prev
    Next
    INDEX
    Explanations

    email/password/name fields

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

    The neuron activates on numeric literals (digits or numbers) in the code or text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    whitespace indentation at the beginning of lines in code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    programming syntax separators and structural tokens in code snippets, especially punctuation that delimits lists, attributes, or tags.

    oai_token-act-pair · gpt-5Triggered 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
     at
    -1.08
     should
    -0.96
    /;
    -0.96
    '';
    -0.94
     but
    -0.93
    oreille
    -0.91
    ."/
    -0.91
    However
    -0.90
     väga
    -0.89
     kuitenkin
    -0.88
    POSITIVE LOGITS
     aparcamiento
    1.02
     estacionamiento
    0.99
    傒
    0.98
     dehors
    0.96
     consiste
    0.96
     forder
    0.94
    奩
    0.91
     origini
    0.90
    后果
    0.90
     Aceite
    0.89
    Activations Density 0.012%

    No Known Activations