© 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. 34-GEMMASCOPE-RES-131K
    7. 4091
    Prev
    Next
    INDEX
    Explanations

    there will be discussion

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

    The neuron reliably fires on numeric list or section markers (e.g. “1.”, “2.”, “3.”, “Minutes from …”), identifying numbered headings or bullet points.

    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
    不僅
    -1.18
     BOTH
    -1.17
    decla
    -1.16
    tells
    -1.15
    declared
    -1.14
    あなたの
    -1.10
    reve
    -1.09
    检验
    -1.09
     ساخته
    -1.09
     เปิด
    -1.09
    POSITIVE LOGITS
     discussed
    1.98
     discussion
    1.88
     discuss
    1.73
     regarding
    1.51
     they
    1.50
     will
    1.48
     there
    1.45
     discussing
    1.45
     everyone
    1.44
     discusión
    1.43
    Activations Density 0.052%

    No Known Activations