© 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. 6651
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to the very first token in each text segment, effectively marking segment boundaries.

    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
    ῆ
    0.96
     그다음에
    0.84
     سپس
    0.77
     多く
    0.76
    ICK
    0.73
    useppe
    0.72
    随后
    0.72
     そして
    0.71
     laterales
    0.70
    
    0.70
    POSITIVE LOGITS
    ب
    1.04
     unanswered
    0.84
     ঘুম
    0.76
     class
    0.75
    mtr
    0.73
     unmet
    0.73
     DevOps
    0.72
    ی
    0.71
    class
    0.71
     superhero
    0.71
    Activations Density 0.000%

    No Known Activations