© 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. 129281
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is pick out technical or domain-specific nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    various specific topics

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     निम्
    -0.98
     ו
    -0.94
     ingresar
    -0.93
    (&$
    -0.89
    这就是
    -0.89
    óstico
    -0.89
     estudian
    -0.88
     ナ
    -0.87
    “
    -0.86
     contento
    -0.86
    POSITIVE LOGITS
     some
    0.98
     vín
    0.97
     itself
    0.97
     membagikan
    0.97
     or
    0.97
     було
    0.95
    愜
    0.93
     Luglio
    0.92
     vertelt
    0.90
     alike
    0.89
    Activations Density 0.169%

    No Known Activations