© 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. 11306
    Prev
    Next
    INDEX
    Explanations

    Swift code syntax

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

    opening parentheses in function calls and initializers, particularly in Swift code.

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

    The neuron fires most strongly on numeric literals (integer and floating‐point constants) in code.

    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
    té
    -1.55
     assise
    -1.53
     öz
    -1.52
    鹟
    -1.52
     pousse
    -1.50
     armée
    -1.48
     lú
    -1.47
     pode
    -1.46
    话音
    -1.46
     Teléfono
    -1.45
    POSITIVE LOGITS
    '
    1.66
     [''],
    1.56
    笮
    1.51
    ykjav
    1.43
     BRARY
    1.42
    寕
    1.41
    栻
    1.41
     of
    1.40
    pertory
    1.39
     akzeptieren
    1.38
    Activations Density 0.012%

    No Known Activations