© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 58902
    Prev
    Next
    INDEX
    Explanations

    disambiguation pages and code constructs

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

    The neuron strongly activates on standalone metadata‐style tokens—e.g. words inside quotes or parentheses like “omitempty,” “unfinished,” “disambiguation,” or “instancetype”—that label or annotate fields rather than ordinary prose.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     =
    -2.31
     ardent
    -2.22
    C
    -2.20
    L
    -2.20
    section
    -2.13
    In
    -2.03
    どのような
    -2.02
     {
    -2.02
    An
    -1.97
     وإ
    -1.95
    POSITIVE LOGITS
    omitempty
    2.91
    ญี่ป
    2.48
    衚
    2.47
    灩
    2.38
    鈇
    2.38
     verwijzen
    2.34
    芣
    2.25
     cérami
    2.22
    งิน
    2.20
    踅
    2.20
    Activations Density 0.001%

    No Known Activations