© 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. 24-GEMMASCOPE-2-RES-16K
    7. 14989
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on the very first substantive word at the start of each new text segment (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
    ی
    1.09
    a
    0.96
    tar
    0.95
    ați
    0.86
    έλ
    0.83
    ా
    0.82
    ા
    0.81
    sType
    0.77
    s
    0.77
    ه
    0.76
    POSITIVE LOGITS
     कहता
    0.84
     thách
    0.83
    不堪
    0.80
    সঙ্গত
    0.80
     объяс
    0.80
     специалисты
    0.79
     рассказыва
    0.78
    ้อย
    0.77
     объеди
    0.77
     também
    0.76
    Activations Density 0.000%

    No Known Activations