© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33408
    Prev
    Next
    INDEX
    Explanations

    programming, mathematical, or statistical questions

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

    The neuron is keyed to generic type‐parameter syntax in code (e.g. uppercase type names like “T” in angle brackets and accompanying `where`/constraint clauses).

    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
    頂いた
    -1.09
     نمایند
    -1.08
     insumos
    -1.03
    ಈ
    -1.00
    åde
    -1.00
     sustancia
    -0.99
     چهارم
    -0.96
     חיצוני
    -0.96
     nestled
    -0.95
     soñar
    -0.95
    POSITIVE LOGITS
    Bjs
    1.06
     ۱۵
    1.04
    nejšie
    1.03
     ۱۶
    1.01
     !!!!
    1.01
     reiter
    1.00
    などの
    0.99
     later
    0.99
     exuber
    0.97
    ństwo
    0.96
    Activations Density 0.002%

    No Known Activations