© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49068
    Prev
    Next
    INDEX
    Explanations

    price, views, meeting, movie

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

    The neuron fires on literal data values—especially standalone booleans and numbers—marking tokens that represent raw constant values in code or statistical/output counts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     cintura
    -0.85
    НИК
    -0.84
     modified
    -0.83
    findOne
    -0.82
    修改
    -0.80
     имени
    -0.79
     Mauer
    -0.75
     modifica
    -0.75
    modified
    -0.74
    chak
    -0.72
    POSITIVE LOGITS
    cheme
    0.82
    ekš
    0.80
     обзор
    0.80
    ↨
    0.79
    丛
    0.77
     galite
    0.75
     память
    0.75
    ‰
    0.74
     fixer
    0.74
     soit
    0.72
    Activations Density 0.002%

    No Known Activations