© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 50387
    Prev
    Next
    INDEX
    Explanations

    entities and their identifiers

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

    The neuron is tuned to pick out tokens consisting of digits (i.e. numeric literals or identifiers containing 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
    ått
    -1.09
    และ
    -1.04
    siye
    -1.01
    หรือ
    -0.92
    rits
    -0.90
    Caprio
    -0.88
    uminum
    -0.88
    tification
    -0.87
    межу
    -0.86
     }^{
    -0.86
    POSITIVE LOGITS
     name
    1.27
    name
    1.09
    Desc
    0.94
     Both
    0.94
    Both
    0.93
    _
    0.93
    desc
    0.92
    Misc
    0.92
    自带
    0.90
    当時は
    0.90
    Activations Density 0.016%

    No Known Activations