© 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. 2384
    Prev
    Next
    INDEX
    Explanations

    The neuron fires most strongly on uncommon, content‐bearing words (significant nouns, verbs, and adjectives) that carry high semantic importance.

    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
    肯定是
    0.50
     சூழ்
    0.49
    umlahan
    0.49
    目安
    0.49
    時点で
    0.49
     الوضع
    0.48
     ቡ
    0.48
     ഹോ
    0.46
     भगत
    0.46
     상황
    0.46
    POSITIVE LOGITS
     forgotten
    2.88
     overlooked
    2.69
     unnoticed
    2.69
     neglected
    2.55
    forgot
    2.30
     Forgotten
    2.30
    Forgotten
    2.30
     ignored
    2.17
     unseen
    2.13
     hidden
    2.09
    Activations Density 1.137%

    No Known Activations