© 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. 28119
    Prev
    Next
    INDEX
    Explanations

    common text sequences

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

    The neuron is highly sensitive to numeric tokens—measurements, list indices, percentages, and other digit‐containing references.

    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
    uresti
    -0.92
     miljoner
    -0.90
     they
    -0.84
     senaste
    -0.84
     gånger
    -0.82
     millioner
    -0.81
    nae
    -0.81
     theyre
    -0.81
     thier
    -0.79
    Liver
    -0.77
    POSITIVE LOGITS
    相変わらず
    0.96
     aspi
    0.95
     inability
    0.94
     フィギュア
    0.94
    你觉得
    0.93
     ningún
    0.88
    漚
    0.85
     نيوز
    0.85
     otomatig
    0.84
    よりも
    0.83
    Activations Density 0.047%

    No Known Activations