© 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. 26523
    Prev
    Next
    INDEX
    Explanations

    code includes and definitions

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

    The neuron flags tokens that contain numbers (digits), i.e. numeric literals or numeric substrings.

    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
     either
    -1.22
     should
    -1.14
     probablemente
    -1.05
     tampoco
    -1.03
     dårlig
    -1.02
     todavía
    -1.02
     before
    -0.99
     in
    -0.96
     outside
    -0.91
     from
    -0.91
    POSITIVE LOGITS
    windig
    1.17
     pasaj
    1.05
     fokus
    1.02
     flotte
    1.01
    嚀
    1.00
     caricatura
    1.00
    zös
    0.98
     moderne
    0.97
     تاکید
    0.96
    your
    0.94
    Activations Density 0.007%

    No Known Activations