© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49237
    Prev
    Next
    INDEX
    Explanations

    district court citations

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

    The neuron fires on sequences of digits and docket‐style numbers, i.e. legal case or filing reference 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
     esfuerzan
    -1.02
    
    -0.97
     Fähig
    -0.94
     riveting
    -0.93
    identes
    -0.90
    aniem
    -0.89
     pelaksanaan
    -0.87
     nacido
    -0.87
     misst
    -0.86
    米国
    -0.85
    POSITIVE LOGITS
    ängt
    0.90
     karakter
    0.90
    Ḍ
    0.88
     WORLD
    0.88
    ↵
    0.88
    jne
    0.86
     الشخص
    0.86
     клиента
    0.85
    herst
    0.80
    stads
    0.78
    Activations Density 0.071%

    No Known Activations