© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 38797
    Prev
    Next
    INDEX
    Explanations

    Pickle Bottom bar Real Industries

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

    The neuron fires on programming identifier tokens—i.e. variable, function, class, and other code symbol names rather than ordinary English words.

    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
    шный
    -1.40
    u
    -1.30
     bancada
    -1.25
     again
    -1.25
    A
    -1.24
     garagem
    -1.23
     iluminação
    -1.20
    它
    -1.20
    _
    -1.20
    更加
    -1.18
    POSITIVE LOGITS
    фи
    1.49
    deinit
    1.32
    🔠
    1.30
    单纯
    1.29
     utan
    1.28
    idavit
    1.27
    ৳
    1.26
    坶
    1.26
    әһ
    1.25
    FormUrlEncoded
    1.24
    Activations Density 0.158%

    No Known Activations