© 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. 22-GEMMASCOPE-RES-131K
    7. 65697
    Prev
    Next
    INDEX
    Explanations

    programming identifiers

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

    The neuron fires strongly on uppercase identifiers—i.e. class, type, or constructor names (capitalized tokens) 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_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
    annulation
    -1.40
    Leírás
    -1.05
    孰
    -1.04
    acruz
    -1.03
     reconoce
    -1.02
    感じが
    -1.02
    conut
    -1.00
    忡
    -1.00
    <mask>
    -0.99
     של
    -0.99
    POSITIVE LOGITS
    トゥ
    1.30
    焢
    1.22
     eccl
    1.20
    Еще
    1.10
     Coruña
    1.09
     katk
    1.07
    ācija
    1.05
    そういえば
    1.05
    Моя
    1.02
     extenso
    1.02
    Activations Density 0.015%

    No Known Activations