© 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. 12655
    Prev
    Next
    INDEX
    Explanations

    job role or position

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

    The neuron fires on the standout content word that introduces each new passage or section, highlighting the key topic token (often the last word of the opening sentence).

    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
    s
    0.96
     та
    0.79
    0.78
     ре
    0.77
     при
    0.77
    t
    0.77
    ты
    0.75
    ы
    0.75
     ח
    0.74
     când
    0.73
    POSITIVE LOGITS
    ینس
    0.85
     Проци
    0.78
     postId
    0.77
    सनीय
    0.73
     denominator
    0.72
    াজ্ঞ
    0.71
    //=
    0.71
    ىڭ
    0.71
    耧
    0.70
    闸
    0.68
    Activations Density 0.001%

    No Known Activations