© 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. 26648
    Prev
    Next
    INDEX
    Explanations

    introduction Bob Dylan

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

    The neuron strongly activates on punctuation marks—especially commas—in 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
    ַּ
    -1.64
    ִּ
    -1.50
    ֶּ
    -1.48
     Кубка
    -1.30
    ּוֹ
    -1.27
    ֵּ
    -1.23
    nữ
    -1.23
     турни
    -1.12
     campeonato
    -1.12
    ָּ
    -1.11
    POSITIVE LOGITS
     you
    1.23
     your
    1.14
    ७
    1.02
     envisioned
    1.01
     modelleri
    0.99
    ὕ
    0.98
     envisaged
    0.98
     prostřed
    0.97
     víc
    0.97
    ff
    0.95
    Activations Density 0.003%

    No Known Activations