© 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. 75602
    Prev
    Next
    INDEX
    Explanations

    code constructs and operators

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

    This neuron detects identifiers used as the object side of a member‐access (the token immediately before a dot) 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
     to
    -3.05
     of
    -2.16
    </em>
    -2.05
     about
    -1.86
    都不能
    -1.80
    greenrobot
    -1.80
    自体
    -1.77
     They
    -1.77
     no
    -1.75
    ↵
    -1.73
    POSITIVE LOGITS
    _
    2.78
    after
    2.36
    軼
    2.31
    źdz
    2.30
    [
    2.20
    Karena
    2.16
    After
    2.13
    媖
    2.13
    ająca
    2.11
    I
    2.09
    Activations Density 0.011%

    No Known Activations