© 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. 25329
    Prev
    Next
    INDEX
    Explanations

    date and time parsing

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

    This neuron fires on date‐related field names in code or database schemas.

    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
     عبار
    -0.96
     domesti
    -0.94
    で
    -0.90
     cioccolato
    -0.89
    では
    -0.88
    *
    -0.88
    ことは
    -0.88
    Ever
    -0.87
     Ouvrez
    -0.87
     Володи
    -0.85
    POSITIVE LOGITS
     date
    1.45
     time
    1.38
     earliest
    1.34
    ksamkeit
    1.17
    šanas
    1.12
    reactivex
    1.11
     strtotime
    1.10
     anderen
    1.08
     Date
    1.07
    的时间
    1.05
    Activations Density 0.017%

    No Known Activations