© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6411
    Prev
    Next
    INDEX
    Explanations

    code comments after punctuation

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

    The neuron is looking for structured enumerations or labels—e.g. section/figure numbers, clause identifiers, schedule and paragraph labels, and other formal item numbers in the text.

    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_31_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
    ↵
    0.66
     either
    0.54
    ‌ನ
    0.52
    <0x0D>
    0.51
     zł
    0.50
     kalın
    0.48
     metal
    0.48
    <0xE2>
    0.47
     metall
    0.46
     mostly
    0.46
    POSITIVE LOGITS
    JSPA
    0.60
    ῴ
    0.60
    ánica
    0.59
     Министерство
    0.59
    ándole
    0.59
    了一条
    0.59
    onDestroy
    0.58
    álním
    0.58
     Анаста
    0.58
    了一個
    0.58
    Activations Density 0.012%

    No Known Activations