© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41908
    Prev
    Next
    INDEX
    Explanations

    system output phrases

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

    The neuron activates on numeric tokens, identifying numbers in the text.

    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
     those
    -2.25
     but
    -1.86
     on
    -1.82
     that
    -1.75
     նաև
    -1.69
     employees
    -1.66
    那些
    -1.59
     Griechenland
    -1.59
     also
    -1.58
     warten
    -1.57
    POSITIVE LOGITS
    ensivo
    1.86
    datable
    1.84
    zers
    1.71
     francisco
    1.70
    îtres
    1.66
    ageable
    1.64
    ǃ
    1.62
    ====
    1.61
    𝑎
    1.59
     oud
    1.58
    Activations Density 0.013%

    No Known Activations