© 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. 34-GEMMASCOPE-RES-131K
    7. 90895
    Prev
    Next
    INDEX
    Explanations

    admitted to the bar

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

    The neuron strongly activates on multi‐digit numeric tokens—dates, years, citation numbers, reporter volumes/pages, and bar or docket registration 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
    monės
    -0.93
    awm
    -0.75
     enough
    -0.72
     voci
    -0.71
    aww
    -0.70
    voi
    -0.69
    nú
    -0.68
    ilihan
    -0.68
    怎样
    -0.67
     улуч
    -0.67
    POSITIVE LOGITS
     admitted
    2.53
     admission
    2.34
    Admission
    2.03
     Admission
    2.02
     admit
    1.98
    admission
    1.86
     admissions
    1.77
     admitting
    1.77
     admittance
    1.76
     admits
    1.63
    Activations Density 0.066%

    No Known Activations