© 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. 50251
    Prev
    Next
    INDEX
    Explanations

    existing INC entities

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

    The neuron detects legal‐case citation metadata (party names, “v.” separators, reporter references, docket numbers, court names, and dates).

    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
    кован
    -1.01
     such
    -0.87
     chromos
    -0.79
    capsulation
    -0.76
    oxin
    -0.76
    Зем
    -0.74
    ();//
    -0.74
    parseFrom
    -0.74
    戮
    -0.73
    共产党
    -0.73
    POSITIVE LOGITS
     existing
    0.94
     Existing
    0.86
    inder
    0.79
    Existing
    0.79
    納
    0.79
     EXISTING
    0.78
    縮
    0.77
    ﻰ
    0.77
     undermine
    0.74
     לעשות
    0.72
    Activations Density 0.010%

    No Known Activations