© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99254
    Prev
    Next
    INDEX
    Explanations

    numbers and patent numbers

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

    The neuron activates on multi‐digit numeric tokens—especially contiguous digit sequences like phone numbers or patent/publication 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_34/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
     Appraisal
    -0.77
    Feedback
    -0.73
    𝙸
    -0.73
     Engen
    -0.72
    カメラ
    -0.71
    斯特
    -0.69
    +-
    -0.68
    対応
    -0.68
    EDIT
    -0.68
    遊戯
    -0.67
    POSITIVE LOGITS
    iaomi
    0.82
     longest
    0.80
    longest
    0.77
    Longest
    0.72
    breve
    0.71
    phenols
    0.71
     snowflakes
    0.71
    ுக
    0.70
    արդ
    0.70
    hhhh
    0.69
    Activations Density 0.014%

    No Known Activations