© 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. 3708
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on the very first content word that begins a new sentence or paragraph.

    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
    as
    0.86
    ot
    0.86
    ის
    0.81
    swe
    0.81
    ian
    0.79
    sp
    0.77
    uction
    0.77
    ს
    0.77
     dick
    0.77
    st
    0.76
    POSITIVE LOGITS
    CriteriaUtils
    0.78
    قل
    0.75
    ﻭ
    0.73
    ORMAL
    0.72
     чемпионате
    0.72
     champions
    0.70
    pdbonly
    0.70
    ﻡ
    0.69
    黾
    0.68
    今後
    0.67
    Activations Density 0.001%

    No Known Activations