© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33267
    Prev
    Next
    INDEX
    Explanations

    code formatting or technical terms

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

    The neuron fires on numeric literals (numbers) in code.

    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
     their
    -1.83
     for
    -1.53
     this
    -1.38
     only
    -1.32
     both
    -1.30
     several
    -1.26
     one
    -1.23
     during
    -1.22
     my
    -1.20
     best
    -1.20
    POSITIVE LOGITS
    onsored
    1.45
    glected
    1.38
     カジュアル
    1.35
     executar
    1.33
     alcuni
    1.33
    argli
    1.32
     alguns
    1.29
     craz
    1.28
    とろ
    1.25
     élar
    1.25
    Activations Density 0.001%

    No Known Activations