© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 32569
    Prev
    Next
    INDEX
    Explanations

    code syntax and structure

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

    The neuron fires on technical identifiers—things like code symbols, module or package names, URLs, and other programming‐related tokens.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    macam
    -0.83
    Click
    -0.83
    Specifically
    -0.82
    Sincerely
    -0.81
     počas
    -0.80
    土耳其
    -0.80
     There
    -0.80
    ırken
    -0.79
    มี
    -0.79
    encana
    -0.79
    POSITIVE LOGITS
    üht
    0.87
     nuovo
    0.86
     lovely
    0.86
    新たに
    0.85
    我还是
    0.85
    垓
    0.85
    安く
    0.84
     Dichtung
    0.83
     inicia
    0.82
     juega
    0.82
    Activations Density 0.114%

    No Known Activations