© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 101597
    Prev
    Next
    INDEX
    Explanations

    class definitions start

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

    class declarations in object-oriented programming code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron is highly responsive to identifiers of classes or modules in type or class definitions (e.g. superclass and class names).

    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
     is
    -1.78
    Ive
    -1.51
    ing
    -1.51
     diversos
    -1.47
    ter
    -1.45
    re
    -1.41
    ive
    -1.39
    ner
    -1.38
    zegovina
    -1.38
    av
    -1.37
    POSITIVE LOGITS
     with
    1.92
     and
    1.41
     latter
    1.38
     competición
    1.38
     meses
    1.30
    venske
    1.30
     uporab
    1.29
     on
    1.29
    If
    1.28
     mesto
    1.27
    Activations Density 0.041%

    No Known Activations