© 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. 72825
    Prev
    Next
    INDEX
    Explanations

    code includes and pragmas

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

    The neuron strongly activates on numeric literals (numbers) in code or markup.

    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
    lizes
    -1.22
    尔夫
    -1.21
    tistic
    -1.09
    neling
    -1.09
    opie
    -1.08
    aneity
    -1.05
    bbing
    -1.04
    riceps
    -1.02
    ofing
    -1.02
     conferma
    -1.02
    POSITIVE LOGITS
     /**
    1.06
    理
    0.93
    本書
    0.91
    ―
    0.91
    京都
    0.91
    १
    0.88
     therewith
    0.87
     irres
    0.86
    ந
    0.86
     repug
    0.85
    Activations Density 0.012%

    No Known Activations