© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 79246
    Prev
    Next
    INDEX
    Explanations

    math terms and concepts

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

    The neuron strongly activates on technical mathematical terminology and notation (e.g. names of structures, symbols, variables, and formula‐related tokens).

    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
     hud
    -0.75
     سخت
    -0.75
     denso
    -0.73
     TIR
    -0.73
     Zona
    -0.73
    拉克
    -0.72
     KOT
    -0.71
    DIN
    -0.71
    PARTY
    -0.71
    ティン
    -0.71
    POSITIVE LOGITS
     перио
    0.85
    polarized
    0.79
    Quaternion
    0.75
     cool
    0.73
     \%,
    0.73
     abelian
    0.72
    Thompson
    0.70
     Eltern
    0.69
    ruvate
    0.69
    Matt
    0.69
    Activations Density 0.031%

    No Known Activations