© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67552
    Prev
    Next
    INDEX
    Explanations

    correlation, product, vectors, norm

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

    This neuron detects mathematical notation tokens—single-letter variables, numbers, and formula symbols in LaTeX/math contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     there
    -0.96
     provides
    -0.88
     on
    -0.87
     provide
    -0.87
     elaborado
    -0.86
    optarg
    -0.84
     handed
    -0.82
     can
    -0.82
     or
    -0.82
    触
    -0.82
    POSITIVE LOGITS
     sfera
    1.05
     joão
    1.00
     culminated
    0.98
    さてさて
    0.98
     Capita
    0.97
     rezultat
    0.95
     investir
    0.94
    Pavel
    0.94
    そういえば
    0.94
    Australie
    0.93
    Activations Density 0.009%

    No Known Activations