© 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. 34-GEMMASCOPE-RES-131K
    4. 76489
    Prev
    Next
    INDEX
    Explanations

    function calls

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

    The neuron fires on function or method names at the point where they’re being invoked (i.e. identifiers immediately followed by an opening parenthesis).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ‘
    -1.16
     acht
    -1.09
     デニム
    -1.08
     ニット
    -1.07
    arsi
    -1.05
    THAT
    -1.01
     lokale
    -0.99
     the
    -0.96
     astounding
    -0.96
    /***
    
    -0.96
    POSITIVE LOGITS
     with
    1.72
     without
    1.57
     by
    1.43
     even
    1.41
     только
    1.30
     даже
    1.29
     only
    1.28
     from
    1.27
     навіть
    1.25
     тільки
    1.20
    Activations Density 0.005%

    No Known Activations