© 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. 44017
    Prev
    Next
    INDEX
    Explanations

    sql schema keywords like USING gist or DEFAULT CHARACTER

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

    The neuron fires on Ruby keyword‐argument punctuation (colons) and method DSL keywords (like type:, dependent:, inverse_of:, do) indicating it’s spotting named parameters/options in Ruby/Rails 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
     emporio
    -1.61
     hjemmeside
    -1.50
    และ
    -1.47
     requête
    -1.38
     zákaz
    -1.38
     Kø
    -1.38
    illir
    -1.36
    泐
    -1.36
     then
    -1.35
     tivi
    -1.35
    POSITIVE LOGITS
    توضیحات
    1.44
     иногда
    1.29
     M
    1.27
    кл
    1.20
    观点
    1.19
    帐户
    1.19
     being
    1.18
     E
    1.18
     möglicherweise
    1.17
    赔偿
    1.17
    Activations Density 0.005%

    No Known Activations