© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6212
    Prev
    Next
    INDEX
    Explanations

    tab/hab/kab/neb/ab/estab sounds

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

    The neuron strongly activates on sequences of capitalized words that look like personal names (e.g. author surnames in citations).

    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
    ྩ
    0.61
     Societies
    0.60
     Calculates
    0.59
    oro
    0.59
     obliter
    0.58
    𝘱
    0.58
     Lesbian
    0.57
     mơ
    0.57
     Phas
    0.56
     Supervis
    0.55
    POSITIVE LOGITS
    hishek
    0.83
    lished
    0.80
    ernacle
    0.79
    onacci
    0.74
    ruary
    0.73
    ulously
    0.73
    bers
    0.72
    iscuit
    0.72
    购
    0.72
    ber
    0.71
    Activations Density 0.818%

    No Known Activations