© 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. 109152
    Prev
    Next
    INDEX
    Explanations

    dates and places

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

    The neuron fires on tokens that are pure numbers or punctuation sequences (e.g. dates, version numbers, ellipses, “Read More” markers, etc.).

    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
    ooplankton
    -0.86
    inheritDoc
    -0.85
    pero
    -0.83
    viewBox
    -0.83
     SCIENCES
    -0.82
     convertView
    -0.81
     hoffen
    -0.80
    HEP
    -0.79
    ríamos
    -0.79
    estrogen
    -0.77
    POSITIVE LOGITS
    Parking
    0.83
    braz
    0.79
     courir
    0.79
     Kenyans
    0.79
     Alabama
    0.78
    固定
    0.74
     bilgisayar
    0.74
     Parking
    0.74
    ainte
    0.73
    bbc
    0.72
    Activations Density 0.002%

    No Known Activations