© 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. 34-GEMMASCOPE-RES-131K
    4. 53862
    Prev
    Next
    INDEX
    Explanations

    current and specific proper nouns

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

    The neuron detects occurrences of the dot member‐access operator (object or method/property access via “.”) 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_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
     Essex
    -0.80
    ̀nh
    -0.79
     Achievement
    -0.78
     COURT
    -0.77
     CONTINUED
    -0.74
    的地方
    -0.71
    ilik
    -0.70
    umo
    -0.70
    ваемые
    -0.69
    ý
    -0.69
    POSITIVE LOGITS
     current
    1.69
    current
    1.48
     currents
    1.05
    Current
    1.03
    現在の
    0.98
    currents
    0.94
    urrent
    0.92
    Ekster
    0.90
     Current
    0.89
     現在
    0.88
    Activations Density 0.013%

    No Known Activations