© 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. 58522
    Prev
    Next
    INDEX
    Explanations

    url suffixes

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

    The neuron is keyed to the forward‐slash characters that appear as path separators in URLs.

    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
    沕
    -1.20
     cualquiera
    -1.19
     will
    -1.18
     But
    -1.17
     requires
    -1.16
     is
    -1.13
    However
    -1.13
     невозможно
    -1.13
    intă
    -1.11
    有無
    -1.10
    POSITIVE LOGITS
     a
    1.80
     such
    1.58
     oreilles
    1.48
     the
    1.45
    aking
    1.29
     acento
    1.29
     ennemis
    1.24
     mengucap
    1.24
     portes
    1.20
     débuts
    1.20
    Activations Density 0.048%

    No Known Activations