© 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. 58336
    Prev
    Next
    INDEX
    Explanations

    Japanese patent application numbers

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

    The neuron fires on numeric tokens—particularly multi‐digit sequences like patent or section 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_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
    ]</
    -0.80
     Torna
    -0.75
     Byers
    -0.72
    ന
    -0.71
    ặ
    -0.71
    کد
    -0.67
    identifikasi
    -0.67
    酛
    -0.67
     nomadic
    -0.66
     Aditya
    -0.66
    POSITIVE LOGITS
    けない
    0.67
    ランプ
    0.66
     lift
    0.65
     lifted
    0.65
    iot
    0.64
    ugu
    0.64
     lifting
    0.64
     ಎ
    0.63
    Pulling
    0.62
    +#+#
    0.62
    Activations Density 0.092%

    No Known Activations