© 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. 10802
    Prev
    Next
    INDEX
    Explanations

    interrogative pronouns and punctuation

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

    This neuron fires on the very first word of each new comment or paragraph, marking the start of a block of text.

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

    direct address to a person by name, typically at the start of a statement or sentence.

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

    direct address (vocative) constructions, especially names or addressees set off by a following comma.

    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
     DIEGO
    0.66
     ।
    0.65
    ்த
    0.63
    ﻬ
    0.60
    ।
    0.58
    😹
    0.58
     MICHAEL
    0.57
    س
    0.57
    ח
    0.57
     ㅎㅎ
    0.57
    POSITIVE LOGITS
    你需要
    0.77
    你可以
    0.75
    you
    0.71
    你怎么
    0.71
    你在
    0.70
    你
    0.70
    你说
    0.69
    এর
    0.68
    为你
    0.67
    你自己
    0.66
    Activations Density 0.051%

    No Known Activations