© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16475
    Prev
    Next
    INDEX
    Explanations

    phone numbers, addresses, or names

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

    The neuron activates on solitary list‐item numbers (standalone digits used as enumeration markers in lists).

    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
     geschenk
    -1.09
     kantoor
    -1.03
    éton
    -1.03
    読書
    -1.02
     завжди
    -1.01
    Ін
    -1.01
     jugado
    -1.01
     participado
    -1.00
     binnen
    -1.00
     escritores
    -0.99
    POSITIVE LOGITS
     statistics
    1.16
     there
    1.10
    https
    1.09
     خوبه
    1.03
    http
    1.02
     https
    1.02
     current
    1.02
     why
    1.00
    www
    0.99
     ऐसे
    0.97
    Activations Density 0.000%

    No Known Activations