© 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. 11643
    Prev
    Next
    INDEX
    Explanations

    This neuron fires strongly on the first substantive word in each new 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
    ড়ান্ত
    0.73
    不错
    0.70
    ায়ন
    0.70
     حر
    0.68
    不錯
    0.68
    隨
    0.68
    很好
    0.66
    就不
    0.66
    LOTREntity
    0.66
    ParentObject
    0.66
    POSITIVE LOGITS
     estes
    0.76
    વ
    0.76
     elev
    0.75
     essa
    0.75
     निराश
    0.74
    stern
    0.71
    𝖔
    0.71
    а
    0.70
    itats
    0.70
    iam
    0.69
    Activations Density 0.000%

    No Known Activations