© 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. 24-GEMMASCOPE-2-RES-16K
    7. 1528
    Prev
    Next
    INDEX
    Explanations

    This neuron fires strongly on the very first token after each segment start (i.e. it detects sentence‐ or segment‐beginning tokens).

    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_24_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
    ustic
    0.81
     empowers
    0.77
    Czas
    0.76
     unsett
    0.76
    Michele
    0.74
     подой
    0.73
    𝙪
    0.73
    німа
    0.72
    ⎆
    0.72
    梁
    0.72
    POSITIVE LOGITS
    mine
    0.73
    0.72
     indole
    0.71
     ४
    0.70
     i
    0.70
    ించాడు
    0.70
    end
    0.70
    BA
    0.69
     आइसो
    0.69
    again
    0.68
    Activations Density 0.563%

    No Known Activations