© 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. 22-GEMMASCOPE-RES-131K
    7. 14686
    Prev
    Next
    INDEX
    Explanations

    code punctuation and structures

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

    The neuron activates on numeric literals (constants) in code.

    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
     πρ
    -0.89
     Czę
    -0.85
    其中
    -0.84
     לח
    -0.83
    Saluti
    -0.82
    gste
    -0.79
    caling
    -0.78
     secondary
    -0.77
    iries
    -0.76
    excited
    -0.76
    POSITIVE LOGITS
    Both
    0.86
     volunt
    0.81
    absch
    0.80
    賛
    0.79
    heça
    0.77
    جلس
    0.77
     persua
    0.77
     both
    0.74
     vernac
    0.74
    All
    0.73
    Activations Density 0.005%

    No Known Activations