© 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. 11462
    Prev
    Next
    INDEX
    Explanations

    male actors, Black partner, sexually explicit

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

    The neuron strongly activates on the final token just before each segment’s end marker, i.e. it detects end-of-segment boundaries.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    parenthetical asides or supplementary information set off by commas or parentheses.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    function-word and punctuation patterns that signal parentheticals, quoted asides, correlative constructions, and common quantifier phrases.

    oai_token-act-pair · gpt-5Triggered 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
    Recent
    0.88
    recent
    0.82
    some
    0.80
    This
    0.77
    mkdirs
    0.75
    connections
    0.74
     recientes
    0.74
    similar
    0.73
    ských
    0.73
    Theme
    0.73
    POSITIVE LOGITS
     і
    0.92
     mashed
    0.88
    𝓎
    0.88
     sole
    0.85
    вання
    0.81
     ingot
    0.81
     differentiator
    0.80
     ג
    0.80
     WAY
    0.79
     اب
    0.79
    Activations Density 0.000%

    No Known Activations