© 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. 73571
    Prev
    Next
    INDEX
    Explanations

    data processing and delays

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

    The neuron is keyed to spotting time‐ or quantity‐related expressions (e.g. durations, numeric intervals, “up to X hours,” “once a week,” 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
     test
    -0.96
     erneut
    -0.93
    藿
    -0.90
     target
    -0.86
     ขาย
    -0.85
     project
    -0.83
     convocatoria
    -0.82
    pēc
    -0.82
     آید
    -0.82
     Тара
    -0.82
    POSITIVE LOGITS
     transactions
    1.33
     processing
    1.19
     delayed
    1.16
     manually
    1.15
     reporting
    1.14
     transaction
    1.07
     processed
    1.05
    processed
    1.05
    odotus
    1.05
    tifully
    1.03
    Activations Density 0.128%

    No Known Activations