© 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. 13707
    Prev
    Next
    INDEX
    Explanations

    special characters and punctuation

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

    The neuron is tuned to pick out specialized technical or scientific jargon—terms you’d see in mathematical, chemical, or engineering contexts.

    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
    tiglie
    -0.91
    tiera
    -0.88
    monary
    -0.87
    尔夫
    -0.86
     было
    -0.85
     partido
    -0.84
    して下さい
    -0.84
    ℏ
    -0.82
    PHOTOS
    -0.82
     מדי
    -0.81
    POSITIVE LOGITS
    both
    0.94
    requires
    0.88
    槓
    0.85
     $
    0.83
    :!
    0.82
    くなる
    0.82
    $^{\
    0.81
     fácilmente
    0.80
    omm
    0.80
    ":"+
    0.80
    Activations Density 0.083%

    No Known Activations