© 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. 93792
    Prev
    Next
    INDEX
    Explanations

    legal cases

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

    The neuron fires on words and numbers characteristic of formal legal or court‐reporting contexts—e.g. tokens in trial descriptions, case citations, and prosecutorial language.

    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
    ことがある
    -0.78
    攻撃
    -0.78
    धि
    -0.77
    Museo
    -0.77
    >();
    -0.75
    illip
    -0.74
    Resin
    -0.72
    的东西
    -0.71
     будущем
    -0.71
    Métodos
    -0.70
    POSITIVE LOGITS
     case
    7.03
     cases
    6.25
    case
    4.94
    cases
    4.50
     Case
    4.28
    Case
    4.28
     Cases
    4.28
    Cases
    4.13
     caso
    3.73
     casos
    3.64
    Activations Density 0.173%

    No Known Activations