© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 77625
    Prev
    Next
    INDEX
    Explanations

    opinion by

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

    The neuron strongly activates on document metadata markers—especially reference and section identifiers (e.g. citation numbers, table/figure labels, funding info headings).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     foto
    -0.84
     ката
    -0.83
     atual
    -0.83
     âm
    -0.83
    ocuk
    -0.77
     truyện
    -0.76
    Amo
    -0.75
     Фото
    -0.75
    equalTo
    -0.75
     później
    -0.75
    POSITIVE LOGITS
     with
    0.84
     ?,
    0.83
    ・
    0.81
    )">
    0.75
    Slice
    0.75
    rch
    0.74
    ricle
    0.73
    empresa
    0.73
    btnSave
    0.72
     trimester
    0.71
    Activations Density 0.010%

    No Known Activations