© 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. 64916
    Prev
    Next
    INDEX
    Explanations

    or document or tab

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

    The neuron responds primarily to numeric tokens (i.e. sequences of digits).

    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
    आर
    -1.14
     kontinu
    -1.12
    之色
    -1.11
    儛
    -1.08
     geograf
    -1.08
     économiques
    -0.99
     colorés
    -0.98
     ТО
    -0.98
     konflikt
    -0.98
     persil
    -0.97
    POSITIVE LOGITS
    깅
    1.08
    مدیر
    1.07
     zumindest
    1.07
    fordshire
    0.96
    ).
    0.95
    culares
    0.95
    ثمان
    0.93
    -//
    0.93
    ®.
    0.93
     хотя
    0.93
    Activations Density 0.012%

    No Known Activations