© 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. 34-GEMMASCOPE-RES-131K
    7. 59670
    Prev
    Next
    INDEX
    Explanations

    questions about correctness

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

    This neuron detects structural section or segment markers (e.g. “In Part 1,” “In Part 2,” numbered headings, and other section‐label tokens).

    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
    ṣ
    -0.92
     Klub
    -0.88
    coloring
    -0.84
    нон
    -0.80
    mrow
    -0.79
    \<
    -0.79
     Lv
    -0.79
     ajustable
    -0.79
     ★★★★
    -0.78
     Norge
    -0.77
    POSITIVE LOGITS
    estyles
    0.98
    copus
    0.94
     competition
    0.88
     freer
    0.88
    せよ
    0.84
     Commencez
    0.84
     Préférences
    0.84
     coll
    0.84
     memories
    0.83
     Constitución
    0.83
    Activations Density 0.015%

    No Known Activations