© 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. 59680
    Prev
    Next
    INDEX
    Explanations

    scrambled eggs

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

    The neuron fires on numeric tokens and symbols in arithmetic or quantity contexts—that is, it detects numbers, numeric sequences, and related signs (like “-” in subtraction or ranges).

    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
     lied
    -0.82
    ágot
    -0.77
    エット
    -0.74
     Ads
    -0.72
    УП
    -0.71
     Apost
    -0.71
    gancia
    -0.71
     Kohl
    -0.71
    grun
    -0.71
    겟
    -0.70
    POSITIVE LOGITS
     scrambled
    1.70
     eggs
    1.34
     ome
    1.27
     omelette
    1.27
     egg
    1.27
     Ome
    1.27
     scramble
    1.19
    Ome
    1.18
     scrambling
    0.90
    eggs
    0.89
    Activations Density 0.034%

    No Known Activations