© 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. 10-GEMMASCOPE-RES-131K
    7. 7151
    Prev
    Next
    INDEX
    Explanations

    dates in legal documents

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

    The neuron fires on tokens that are part of dates or numerical date references (e.g. numbers in a "Dated: …" line and month/day/year components).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     Usted
    -1.63
    owm
    -1.60
    coj
    -1.54
    idlertid
    -1.47
    が無い
    -1.47
     BECAUSE
    -1.47
    刻意
    -1.46
     Você
    -1.45
     Nuestros
    -1.45
    вушка
    -1.44
    POSITIVE LOGITS
     with
    1.73
     for
    1.51
     one
    1.49
     from
    1.41
    they
    1.38
     got
    1.38
     made
    1.37
    1
    1.34
     comes
    1.34
     takes
    1.31
    Activations Density 0.027%

    No Known Activations