© 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. 15757
    Prev
    Next
    INDEX
    Explanations

    phone numbers

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

    The neuron predominantly activates on digit sequences (especially phone numbers).

    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
     still
    -1.11
     also
    -1.09
     such
    -1.04
     other
    -1.02
    UMBIA
    -0.98
    Their
    -0.96
     mehrere
    -0.96
    ֖
    -0.96
     that
    -0.95
     this
    -0.95
    POSITIVE LOGITS
     or
    1.46
     your
    1.31
    珦
    1.15
     incríveis
    1.14
     YOUR
    1.07
     conç
    1.06
     recarg
    1.06
     skvě
    1.06
    Nazionale
    1.05
    йтинг
    1.05
    Activations Density 0.007%

    No Known Activations