© 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. 82624
    Prev
    Next
    INDEX
    Explanations

    `ref` queries

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

    The neuron activates on long runs of punctuation—especially repeated dashes or similar characters—used as ASCII‐art separators or horizontal rule markers in code comments.

    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
    setSearch
    -0.77
    驾
    -0.74
     typically
    -0.72
     Pach
    -0.72
    Rode
    -0.68
    handleDelete
    -0.68
    arnos
    -0.68
    Inspec
    -0.67
     ouvrages
    -0.66
    ǻ
    -0.66
    POSITIVE LOGITS
     Georgetown
    0.87
    fuels
    0.84
    Blockly
    0.83
    getState
    0.82
    INDEX
    0.82
    player
    0.82
    Index
    0.81
    莎
    0.81
     player
    0.80
    onha
    0.80
    Activations Density 0.000%

    No Known Activations