© 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. 22-GEMMASCOPE-RES-131K
    7. 66445
    Prev
    Next
    INDEX
    Explanations

    connecting opposing concepts

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

    The neuron strongly activates on tokens that are numeric or contain digits (i.e. number-like tokens).

    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
     coverage
    -1.05
    amsmath
    -1.04
     Sitten
    -0.98
     Atem
    -0.97
    elding
    -0.96
    locals
    -0.96
    second
    -0.95
    essive
    -0.94
    your
    -0.94
    downs
    -0.94
    POSITIVE LOGITS
     någon
    1.01
     quando
    0.97
     ebbe
    0.93
     perché
    0.88
    LoadData
    0.85
    LAURA
    0.83
     kuliah
    0.82
     porque
    0.82
     duelo
    0.82
     Reconciliation
    0.81
    Activations Density 0.089%

    No Known Activations