© 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. 63465
    Prev
    Next
    INDEX
    Explanations

    divorce and separation

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

    The neuron primarily fires on numeric tokens—especially dates, years, and other digit sequences.

    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
    tosí
    -0.81
    区间
    -0.79
    ibus
    -0.79
     recreated
    -0.76
    кість
    -0.75
     timbul
    -0.75
     resurrected
    -0.75
     predecess
    -0.73
     condens
    -0.73
    ᱓
    -0.72
    POSITIVE LOGITS
     divorce
    5.41
     divorced
    3.83
     divor
    3.75
     Divorce
    3.64
     separation
    3.61
    Divorce
    3.52
     Separation
    3.02
    離婚
    2.97
     separated
    2.94
    Separation
    2.88
    Activations Density 0.065%

    No Known Activations