© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7881
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on the very first substantive token in each new segment or line (i.e. the first content word after a break).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    u
    0.79
    ad
    0.77
    t
    0.76
    dU
    0.74
    id
    0.72
     automakers
    0.70
    d
    0.69
    ther
    0.67
    tens
    0.67
    r
    0.67
    POSITIVE LOGITS
    лиз
    0.78
    SERVICIO
    0.78
    ючись
    0.76
     जिससे
    0.75
     съм
    0.74
    שרה
    0.73
    ElementString
    0.72
    abria
    0.72
    classAttribute
    0.72
     remar
    0.71
    Activations Density 0.000%

    No Known Activations