© 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. 34-GEMMASCOPE-RES-131K
    7. 106554
    Prev
    Next
    INDEX
    Explanations

    variable names and identifiers

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

    The neuron is detecting 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_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
    hever
    -1.14
    '");
    -1.11
     reisen
    -1.09
    ícias
    -1.08
     ferien
    -1.05
    ousand
    -1.04
     报告
    -1.04
     courseId
    -1.02
    omorphisms
    -1.02
    $?
    -1.02
    POSITIVE LOGITS
     description
    1.36
     phone
    1.33
     gender
    1.31
     email
    1.29
     location
    1.26
     telephone
    1.21
     status
    1.21
     price
    1.19
     name
    1.16
     date
    1.13
    Activations Density 0.010%

    No Known Activations