© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 15-GEMMASCOPE-2-TRANSCODER-262K
    7. 13463
    Prev
    Next
    INDEX
    Explanations

    The provided lists are insufficient to determine a pattern or behavior for the neuron. `MAX_ACTIVATING_TOKENS` and `TOKENS_AFTER_MAX_ACTIVATING_TOKEN` are empty, preventing any analysis based on specific token context. The `TOP_POSITIVE_LOGITS` contain words from different languages (Chinese, Russian, English) and fragments. The `TOP_ACTIVATING_TEXTS` consist of code snippets. Without common keywords, linguistic or structural patterns, or clear contextual links between these elements, it's impossible to derive a meaningful explanation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_15_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     LOS
    0.38
    s
    0.37
     accom
    0.33
     sno
    0.33
    zunehmen
    0.31
    Invest
    0.31
     its
    0.31
    lä
    0.31
     forgiveness
    0.31
     poner
    0.30
    POSITIVE LOGITS
    说过
    0.37
    доро
    0.35
     Chromatogr
    0.34
     Многие
    0.34
    ICIO
    0.34
    marshalO
    0.33
    ພວກເຮ
    0.33
    $，
    0.33
     Эн
    0.33
    我们也
    0.32
    Activations Density 0.001%

    No Known Activations