© 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. 34-GEMMASCOPE-RES-131K
    7. 57455
    Prev
    Next
    INDEX
    Explanations

    police activity

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

    This neuron highlights structured report metadata, especially timestamps and other numeric time‐of‐day notations (e.g., “2 p.m.”, “05:11 a.m.”) in police‐style incident reports.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     murdering
    -0.94
     уби
    -0.83
     fabrik
    -0.80
    acağı
    -0.79
     homicide
    -0.79
     consectetur
    -0.78
    终于
    -0.77
    usaha
    -0.77
     murders
    -0.77
    ोंने
    -0.77
    POSITIVE LOGITS
     Patrol
    1.43
     patrol
    1.37
     responding
    1.26
     officer
    1.22
     officers
    1.22
    patrol
    1.15
    Patrol
    1.08
     investigating
    1.08
     patrols
    1.07
     responded
    1.06
    Activations Density 0.052%

    No Known Activations