© 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. 34647
    Prev
    Next
    INDEX
    Explanations

    Utah and LDS topics

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

    This neuron specializes in spotting legal case captions—party names, the “v.” separator, and status labels like Plaintiffs, Respondents, Defendants, and Appellants.

    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
     NYS
    -0.87
    上帝
    -0.78
     goddamn
    -0.76
     BIBLE
    -0.75
     predeceased
    -0.75
    Laf
    -0.74
     fucked
    -0.74
    APIView
    -0.73
     Middletown
    -0.73
    fucking
    -0.72
    POSITIVE LOGITS
     Utah
    2.75
     BYU
    2.53
    Utah
    2.50
     LDS
    2.14
     Mormon
    1.98
    utah
    1.84
    LDS
    1.82
     Brigham
    1.71
     Provo
    1.55
     utah
    1.49
    Activations Density 0.136%

    No Known Activations