© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 10461
    Prev
    Next
    INDEX
    Explanations

    necessary to

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

    The neuron fires strongly on tokens that occur inside parentheses, i.e. parenthetical asides.

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

    words or phrases that introduce lists, examples, or additional items.

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

    function words that connect or modify clauses and nouns, such as prepositions, relative/possessive markers, auxiliaries, and adverbs indicating degree or inclusion.

    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
     Վ
    0.85
     можуть
    0.76
     inoltre
    0.71
     katanya
    0.71
     щоб
    0.71
     Кар
    0.70
     oraz
    0.70
     которые
    0.68
     LIMITED
    0.68
     possono
    0.68
    POSITIVE LOGITS
    étais
    0.91
    ostasis
    0.82
    भिन्न
    0.80
    웠
    0.77
    Estado
    0.76
     नव्हते
    0.75
    État
    0.75
     Staats
    0.73
    Está
    0.73
    MeOH
    0.71
    Activations Density 0.000%

    No Known Activations