© 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. 58576
    Prev
    Next
    INDEX
    Explanations

    an, en, in, on syllables

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

    This neuron detects numeric tokens that represent years or other four-digit date-like numbers.

    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
    pley
    -0.79
    hopefully
    -0.77
     hope
    -0.77
    彼此
    -0.76
     Otherwise
    -0.75
     pronto
    -0.72
    discover
    -0.69
    écou
    -0.69
     powsta
    -0.69
     JComboBox
    -0.68
    POSITIVE LOGITS
    ními
    0.88
    Diff
    0.85
    avoir
    0.80
    Conductor
    0.75
    ވ
    0.73
    Ori
    0.73
    ůl
    0.73
     Fizz
    0.71
     zile
    0.71
    螨
    0.71
    Activations Density 0.038%

    No Known Activations