© 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. 129387
    Prev
    Next
    INDEX
    Explanations

    The neuron detects numeric tokens that are part of four-digit year mentions (e.g. “2021,” “2023,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    years and dates

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     \
    -2.28
     aimed
    -1.85
    replyreport
    -1.84
    т
    -1.81
    鲑
    -1.80
    \
    -1.77
    包含
    -1.73
    喼
    -1.72
     terdiri
    -1.70
    を書く
    -1.69
    POSITIVE LOGITS
    e
    2.17
     Our
    1.89
    his
    1.88
    ia
    1.85
     варианты
    1.73
     звезды
    1.69
     vorigen
    1.67
    er
    1.64
     Kommission
    1.64
     faciliter
    1.63
    Activations Density 0.001%

    No Known Activations