© 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. 835
    Prev
    Next
    INDEX
    Explanations

    relationships, containment, and membership

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

    The neuron is chiefly firing on numeric tokens and numeric literals (i.e. any standalone digits or numbers) in code‐and‐query examples.

    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
     nicht
    -0.83
     возвра
    -0.80
    ामी
    -0.77
     não
    -0.77
     Fahrrad
    -0.76
     spune
    -0.75
     Store
    -0.75
     оружи
    -0.74
     Marijuana
    -0.73
     zugelassen
    -0.73
    POSITIVE LOGITS
    memberOf
    1.27
     belongs
    1.20
     belonged
    1.18
     belong
    1.17
     containment
    1.12
     contains
    1.10
     has
    1.05
     Has
    1.04
     ISA
    1.03
     Containment
    1.02
    Activations Density 0.055%

    No Known Activations