© 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. 33206
    Prev
    Next
    INDEX
    Explanations

    code and data structures

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

    The neuron strongly activates on numeric literals (e.g. integers, floats, hex values) 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
     there
    -0.99
    There
    -0.87
    there
    -0.82
     незавер
    -0.82
     which
    -0.81
     anel
    -0.79
     Although
    -0.79
     by
    -0.79
    idade
    -0.79
     jerawat
    -0.78
    POSITIVE LOGITS
    ftes
    1.02
     innumer
    0.91
    Materiál
    0.90
     commencer
    0.89
     particuliers
    0.89
    LLYWOOD
    0.88
    décoration
    0.87
    atzen
    0.87
     THROUGH
    0.86
     világ
    0.85
    Activations Density 0.034%

    No Known Activations