© 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. 64724
    Prev
    Next
    INDEX
    Explanations

    names and email addresses

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

    The neuron fires on any token containing numeric digits (e.g. years, version numbers, or numbers in identifiers).

    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
    mente
    -0.90
    嫻
    -0.87
     time
    -0.86
     group
    -0.82
     only
    -0.81
    𝗛
    -0.81
     الذين
    -0.81
     Cumber
    -0.81
     emperador
    -0.79
    鴻
    -0.79
    POSITIVE LOGITS
     delightfully
    0.93
     оружия
    0.89
     mały
    0.87
    潴
    0.85
    imel
    0.85
    ToObject
    0.85
     wonderfully
    0.85
     stehe
    0.85
    лат
    0.84
     beschreiben
    0.84
    Activations Density 0.071%

    No Known Activations