© 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. 14885
    Prev
    Next
    INDEX
    Explanations

    loop variables and calculations

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

    The neuron is keyed to spotting numeric literals (digits and numbers) in code snippets.

    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
     hadden
    -1.02
    ENODEV
    -0.99
     innego
    -0.93
    চ
    -0.93
     gespeichert
    -0.92
     hará
    -0.92
    רְ
    -0.91
     hace
    -0.89
    מְ
    -0.88
     getItemCount
    -0.87
    POSITIVE LOGITS
     (
    1.02
    ṃ
    1.02
     emballage
    0.96
    $
    0.96
     färg
    0.95
    )}$,
    0.92
     anta
    0.90
    щиков
    0.90
     even
    0.89
    ване
    0.88
    Activations Density 0.015%

    No Known Activations