© 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. 10-GEMMASCOPE-RES-131K
    4. 12609
    Prev
    Next
    INDEX
    Explanations

    declarations and code blocks };

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

    The neuron fires on numeric literals used in code, especially numbers in C‐style initializers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    should
    -2.55
    some
    -2.52
     deliciously
    -2.50
    鶿
    -2.50
     permettant
    -2.47
    酹
    -2.45
    both
    -2.45
     nieuwe
    -2.44
     nouveaux
    -2.44
     delightfully
    -2.42
    POSITIVE LOGITS
    ).
    3.33
     because
    2.95
     if
    2.83
    );
    2.83
    .”
    2.77
    }=\
    2.73
    ');
    2.67
    ))
    2.64
     as
    2.58
     when
    2.48
    Activations Density 0.006%

    No Known Activations