© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5722
    Prev
    Next
    INDEX
    Explanations

    medical treatments or conservative approaches

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron specializes in spotting words immediately before a numbered or parenthesized list item (i.e. the token right before “(1)”, “(2)”, etc.).

    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
     végétaux
    1.01
     communautés
    0.82
     câteva
    0.81
     épars
    0.77
    ō
    0.76
     răz
    0.75
    usalem
    0.75
     nouveaux
    0.75
     chiam
    0.73
     ако
    0.73
    POSITIVE LOGITS
    OVER
    0.86
    เร็จ
    0.83
    references
    0.79
    OSR
    0.77
    number
    0.77
    <unused2140>
    0.76
    fore
    0.74
    limits
    0.73
    ↵↵
    0.72
    name
    0.72
    Activations Density 0.001%

    No Known Activations