© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39617
    Prev
    Next
    INDEX
    Explanations

    Kotlin `val` declarations

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

    The neuron consistently activates on semicolons terminating code declarations (e.g. end-of-line “;” in TypeScript/JavaScript).

    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
     sebut
    -1.78
     ujar
    -1.78
     навіть
    -1.66
     стоковые
    -1.54
    这件事情
    -1.52
     vokal
    -1.52
     meninggal
    -1.50
     praktis
    -1.50
     idyl
    -1.45
     hatta
    -1.45
    POSITIVE LOGITS
    ↵↵
    2.36
     they
    1.93
     that
    1.75
    \
    1.56
     zwłaszcza
    1.38
     gän
    1.37
     zacht
    1.37
    ve
    1.37
    不太
    1.36
     būti
    1.36
    Activations Density 0.032%

    No Known Activations