© 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. 36301
    Prev
    Next
    INDEX
    Explanations

    ' years

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

    This neuron selectively fires on alphanumeric “code‐like” tokens—numbers, all-caps identifiers, and model/version labels.

    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
    “……
    -1.34
    atería
    -1.20
    eara
    -1.13
    iança
    -1.06
    าหล
    -1.05
    Zunanje
    -1.04
    株式会社
    -1.03
     spécialisée
    -1.03
    icolo
    -1.02
     دشمن
    -1.02
    POSITIVE LOGITS
     geladeira
    1.02
     my
    1.00
     ọ
    1.00
     chè
    0.99
    Ár
    0.97
     trinh
    0.93
    بیش
    0.92
    Моя
    0.91
    curator
    0.91
    Etienne
    0.90
    Activations Density 0.013%

    No Known Activations