© 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. 8686
    Prev
    Next
    INDEX
    Explanations

    preparing for tasks

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

    The neuron strongly fires on fixed multi‐word expressions (common collocations or named phrases) rather than individual words.

    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
    t
    0.86
    때
    0.78
     psig
    0.77
    EXPL
    0.77
    le
    0.77
    m
    0.77
    싸
    0.77
    mak
    0.76
    GATE
    0.75
    ेल
    0.75
    POSITIVE LOGITS
    ដូច
    0.81
     Lösung
    0.78
     Folgen
    0.77
     ۾
    0.76
     gemeins
    0.73
    اديم
    0.73
    ជីវ
    0.71
     Apesar
    0.71
     وبر
    0.70
     Dla
    0.70
    Activations Density 0.000%

    No Known Activations