© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32109
    Prev
    Next
    INDEX
    Explanations

    ending in able or proper nouns

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

    The neuron responds to Java source‐code structural keywords and symbols—especially access modifiers (like “public,” “static,” “final”) and block delimiters (braces).

    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
    𝗽
    -0.79
    %#
    -0.77
     éter
    -0.76
    vány
    -0.75
    +
    -0.75
    іе
    -0.75
    澳门
    -0.75
    民主
    -0.74
     without
    -0.74
    恫
    -0.73
    POSITIVE LOGITS
    caud
    0.93
    fita
    0.87
    vron
    0.87
     plau
    0.85
    องค์
    0.82
    ⑬
    0.82
     Dinas
    0.80
     Advocates
    0.80
    colorPrimary
    0.80
    Thanos
    0.80
    Activations Density 0.001%

    No Known Activations