© 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. 58626
    Prev
    Next
    INDEX
    Explanations

    something followed by punctuation

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

    The neuron activates on vague placeholder terms—especially the word “something” (and its variants like “similar”).

    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
    ൭
    -1.63
    ਲ਼
    -1.59
    䥑
    -1.56
    萏
    -1.55
     vlastnosti
    -1.54
     ส์
    -1.52
    UALA
    -1.51
    䜣
    -1.50
    訁
    -1.49
     ܗ
    -1.48
    POSITIVE LOGITS
     of
    2.42
    .
    2.08
     from
    1.97
     increase
    1.87
     when
    1.84
     like
    1.71
     extensive
    1.60
     because
    1.60
     other
    1.59
     또는
    1.57
    Activations Density 0.013%

    No Known Activations