© 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. 14392
    Prev
    Next
    INDEX
    Explanations

    declaring or announcing

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

    This neuron picks out narrative‐style/book‐blurb language—especially words signaling a historical or biographical “tale” or “story” (e.g. tale, story, illustrations, detailed, inspiring).

    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
    很多
    0.89
     many
    0.81
    有很多
    0.80
     sommige
    0.80
     I
    0.80
     There
    0.79
     soms
    0.77
     These
    0.76
    某些
    0.75
     we
    0.74
    POSITIVE LOGITS
     พร้อม
    0.78
    ኾ
    0.75
     заявил
    0.70
    үкт
    0.70
     ประก
    0.66
    প্রতিষ্ঠ
    0.65
     waarmee
    0.64
    ประกาศ
    0.63
    を迎
    0.63
     frontman
    0.63
    Activations Density 13.405%

    No Known Activations