© 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. 130680
    Prev
    Next
    INDEX
    Explanations

    The neuron responds to isolated numeric tokens (lines or fields consisting solely of numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    sequences of C and 0

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     gyóg
    -0.97
    сете
    -0.94
    язы
    -0.90
     atleta
    -0.89
    пет
    -0.88
     ziem
    -0.86
     Teknologi
    -0.86
     fiables
    -0.84
    пой
    -0.84
     dentaire
    -0.84
    POSITIVE LOGITS
    qiang
    0.92
    Ayuda
    0.90
    协助
    0.87
     اینجا
    0.87
     DuPont
    0.85
     gorgeous
    0.85
     bénéfic
    0.84
    苼
    0.82
     zako
    0.82
     solchen
    0.82
    Activations Density 0.039%

    No Known Activations