© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71745
    Prev
    Next
    INDEX
    Explanations

    resetting or clearing code

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

    This neuron detects code‐style punctuation tokens—especially semicolons, parentheses, and 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_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
     Secondo
    -0.94
    WOR
    -0.94
    ях
    -0.89
    県道
    -0.89
     パソコン
    -0.88
     troopers
    -0.87
    țiune
    -0.87
    离开
    -0.87
     nghị
    -0.86
     permanence
    -0.84
    POSITIVE LOGITS
     reset
    0.95
     it
    0.89
     fresh
    0.87
     empty
    0.84
     everything
    0.84
     cleared
    0.81
    ța
    0.79
    텅
    0.78
    ݣ
    0.78
    giarism
    0.77
    Activations Density 0.003%

    No Known Activations