© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 32878
    Prev
    Next
    INDEX
    Explanations

    times and days of week

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

    The neuron is specialized to pick out time expressions (numeric times of day and surrounding cues like “around,” clock times, etc.).

    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
     otel
    -1.21
     figlio
    -1.16
     angin
    -1.14
     tille
    -1.14
     for
    -1.14
     campione
    -1.13
     stiker
    -1.13
    瞩
    -1.13
     riks
    -1.11
    tieth
    -1.10
    POSITIVE LOGITS
     Thursday
    1.22
     afternoon
    1.16
     evening
    1.13
    extrême
    1.10
     descrito
    1.10
     niebies
    1.08
     yesterday
    1.08
    時間に
    1.07
     Tuesday
    1.06
     Sunday
    1.05
    Activations Density 0.010%

    No Known Activations