© 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. 22-GEMMASCOPE-RES-131K
    7. 14509
    Prev
    Next
    INDEX
    Explanations

    without replacement

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

    The neuron fires on numeric tokens—especially multi‐digit numbers and decimals—i.e. it’s detecting standalone numeric values in the 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_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
    create
    -0.84
    ccion
    -0.78
    すみません
    -0.78
    >"+
    -0.78
     🔥
    -0.78
    DCs
    -0.77
    首先
    -0.77
    кти
    -0.77
    roben
    -0.76
    SuspendLayout
    -0.75
    POSITIVE LOGITS
     beide
    1.16
     BOTH
    1.07
     both
    0.99
    どちら
    0.96
    anso
    0.95
     تاری
    0.92
    これが
    0.90
     czerwony
    0.90
     šed
    0.89
    Muffin
    0.89
    Activations Density 0.004%

    No Known Activations