© 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. 129010
    Prev
    Next
    INDEX
    Explanations

    This neuron detects numeric tokens that are part of timestamp strings (hours, minutes, seconds) in generated-by–javadoc comments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    central auditory processing

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     just
    -1.02
     toiminta
    -0.94
     various
    -0.93
     many
    -0.88
     continuously
    -0.86
     constantly
    -0.86
     their
    -0.85
    еве
    -0.85
     then
    -0.84
    ariam
    -0.84
    POSITIVE LOGITS
     hecta
    0.94
    pog
    0.93
     meninos
    0.88
    millions
    0.86
    ピック
    0.85
    learned
    0.85
     Telescope
    0.85
     professores
    0.83
     рели
    0.82
    zension
    0.82
    Activations Density 0.003%

    No Known Activations