© 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. 34-GEMMASCOPE-RES-131K
    7. 93628
    Prev
    Next
    INDEX
    Explanations

    warped product spaces

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

    The neuron fires on numeric tokens—especially larger multi-digit numbers—within scholarly text.

    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
     Paleo
    -0.88
    hicle
    -0.85
    EndGlobalSection
    -0.81
    taylor
    -0.77
     miesz
    -0.77
    olished
    -0.75
    setViewportView
    -0.74
     DHA
    -0.73
     uczniów
    -0.73
    *}
    -0.72
    POSITIVE LOGITS
     spaces
    0.94
     geometry
    0.89
     three
    0.86
     complex
    0.82
     space
    0.81
     telescope
    0.79
    ski
    0.79
     six
    0.78
     shrinks
    0.77
     seven
    0.77
    Activations Density 0.033%

    No Known Activations