© 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. 3968
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on the special “blank‐line” or paragraph‐break tokens (i.e. the hidden separator/newline markers between text blocks).

    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
     ਦੀ
    0.73
     mempengaruhi
    0.73
     وهذا
    0.71
     دیکھیں
    0.71
     ਹੀ
    0.71
     року
    0.69
    που
    0.69
    с
    0.69
     식으로
    0.69
     tejidos
    0.68
    POSITIVE LOGITS
    Traité
    0.74
    iya
    0.73
    lja
    0.72
    é
    0.72
    ओ
    0.71
    \(
    0.70
    ivorous
    0.70
    初
    0.68
    半径
    0.67
    リーン
    0.67
    Activations Density 0.000%

    No Known Activations