© 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. 72925
    Prev
    Next
    INDEX
    Explanations

    file path components

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

    The neuron consistently activates on numeric tokens or numeric literals (e.g., standalone numbers or digit sequences).

    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
    agerie
    -1.12
    uring
    -1.11
    ational
    -1.11
    ocene
    -1.08
    tions
    -1.05
    oping
    -1.02
    tocks
    -1.02
    aried
    -1.01
    spective
    -1.00
    ening
    -0.98
    POSITIVE LOGITS
    getT
    1.00
     白菜
    0.98
     població
    0.94
     already
    0.94
     مهند
    0.93
     putea
    0.93
    ſi
    0.93
    મ
    0.92
     ingredi
    0.92
     sulfu
    0.91
    Activations Density 0.021%

    No Known Activations