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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14112
    Prev
    Next
    INDEX
    Explanations

    special characters and punctuation

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

    This neuron responds to isolated punctuation or symbol tokens (HTML entities, ellipses, stray Unicode marks, brackets, 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_10/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
     acrí
    -1.67
     podremos
    -1.63
    哢
    -1.53
     おすすめ
    -1.51
     decorativo
    -1.50
     reciclado
    -1.50
    demment
    -1.49
     contaba
    -1.48
     competitivo
    -1.48
    they
    -1.43
    POSITIVE LOGITS
     in
    1.97
     for
    1.73
     ſame
    1.65
     или
    1.58
    ,
    1.50
    6
    1.49
     his
    1.47
     یا
    1.45
    各样的
    1.41
    3
    1.41
    Activations Density 0.002%

    No Known Activations