© 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. 11418
    Prev
    Next
    INDEX
    Explanations

    florets or generate command

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

    This neuron fires strongly on the very first non‐trivial content word at the start of each new text segment—that is, it detects the initial content word immediately following a segment boundary.

    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
     jūs
    0.73
    Ў
    0.68
    eal
    0.66
    u
    0.65
    idual
    0.64
    <0xBB>
    0.63
    ців
    0.63
     quench
    0.63
    oi
    0.63
     démarche
    0.63
    POSITIVE LOGITS
    iega
    0.83
    बैंड
    0.80
     Elles
    0.75
     생활
    0.75
    את
    0.75
    ਦ
    0.74
     groves
    0.74
     anti
    0.73
     byt
    0.73
     ব্রিটেন
    0.73
    Activations Density 0.000%

    No Known Activations