© 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. 427
    Prev
    Next
    INDEX
    Explanations

    creation date or author

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

    The neuron fires on header-comment metadata fields—especially tokens marking author or creation dates (e.g. “Created,” “date,” and year numbers).

    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
     POETRY
    -1.11
    B
    -1.09
     vět
    -1.06
     JULIO
    -1.05
     PENNSYLVANIA
    -1.02
     一起
    -1.02
     will
    -1.02
    avimo
    -0.96
    がて
    -0.96
     we
    -0.95
    POSITIVE LOGITS
     by
    1.17
     anlam
    1.01
    NEED
    1.01
    IFEST
    1.01
     NEED
    1.01
    cartão
    0.99
    Medida
    0.98
     ceny
    0.98
    ficheiro
    0.97
    鶴
    0.96
    Activations Density 0.008%

    No Known Activations