© 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. 22-GEMMASCOPE-RES-131K
    4. 71571
    Prev
    Next
    INDEX
    Explanations

    code statement terminator

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

    The neuron fires on the semicolon that terminates an import statement—i.e. it detects the end‐of‐line ‘;’ in Java import declarations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    semicolons at the end of import statements in Java code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    Java/Android library type and annotation class names, especially appearing in import or declaration contexts.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    -1.09
    =
    -1.04
    //};
    -1.04
     IDisposable
    -1.02
    GenerationType
    -1.02
     (
    -1.01
     (=
    -1.01
     """",
    -1.00
    																									
    -0.98
    content
    -0.96
    POSITIVE LOGITS
     pratiquement
    1.00
    Grupos
    0.98
     culoare
    0.97
    isGrid
    0.96
     такая
    0.96
     voed
    0.95
    celtic
    0.94
     ROH
    0.94
    fao
    0.93
     berbagai
    0.93
    Activations Density 0.003%

    No Known Activations