© 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. 22-GEMMASCOPE-RES-131K
    7. 84373
    Prev
    Next
    INDEX
    Explanations

    connecting concurrent actions

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

    The neuron detects numbered list markers or section/item numbering (e.g., “(1)”, “(2)”, “(3)”, etc.) in structured academic‐style text.

    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
     trône
    -1.00
    chsel
    -0.97
    囂
    -0.96
    敺
    -0.94
    ֠
    -0.93
    XVIII
    -0.92
    tså
    -0.91
     rése
    -0.91
     etek
    -0.91
     花纹
    -0.90
    POSITIVE LOGITS
     provide
    1.11
     helps
    1.10
     provides
    1.08
     ensure
    1.04
     provided
    1.02
     allow
    0.98
     frequent
    0.96
     enable
    0.94
    更是
    0.92
     added
    0.91
    Activations Density 0.057%

    No Known Activations