© 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. 34-GEMMASCOPE-RES-131K
    7. 63768
    Prev
    Next
    INDEX
    Explanations

    academic paper citations

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

    The neuron fires on numeric tokens (digit sequences such as reference numbers, page numbers, arXiv/DOI identifiers, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     لئے
    -0.85
    Ти
    -0.84
     failure
    -0.77
    Quando
    -0.76
    今まで
    -0.75
     livres
    -0.74
    学的
    -0.73
    esanan
    -0.73
    Cuándo
    -0.73
     Ibidem
    -0.73
    POSITIVE LOGITS
    0
    0.82
     physicists
    0.81
     Tales
    0.77
     obstructive
    0.76
     aussitôt
    0.72
    ０
    0.71
     marketplaces
    0.71
    мулятор
    0.70
     `.
    0.70
    ія
    0.69
    Activations Density 0.007%

    No Known Activations