© 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. 34484
    Prev
    Next
    INDEX
    Explanations

    text and following punctuation

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

    The neuron fires whenever it encounters the literal token “text,” as used in code, markup, or style declarations.

    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
    ⵓ
    -0.79
    線が
    -0.76
    持续
    -0.76
    国道
    -0.75
     منذ
    -0.71
     Nutri
    -0.71
     incentiv
    -0.71
    于
    -0.71
    依旧
    -0.70
     生命
    -0.70
    POSITIVE LOGITS
     text
    0.94
     vaiz
    0.84
    ured
    0.82
    Sì
    0.82
    hinh
    0.81
    ulties
    0.81
    火的
    0.81
     decent
    0.78
    elesen
    0.78
    דים
    0.77
    Activations Density 0.017%

    No Known Activations