© 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. 59424
    Prev
    Next
    INDEX
    Explanations

    legal citations and state names

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

    The neuron fires on uppercase or numeric tokens typically found in formal document headings—dates, section numbers, and all-caps titles.

    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
     spied
    -0.77
     új
    -0.77
     Oregon
    -0.75
     arbets
    -0.73
    uyan
    -0.72
     Dun
    -0.71
     Worth
    -0.71
    🏉
    -0.71
     สุ
    -0.71
     werken
    -0.70
    POSITIVE LOGITS
     tranquill
    0.85
     essenziale
    0.76
    ﻩ
    0.74
    ulée
    0.74
    decomposition
    0.72
     Gabb
    0.71
    调
    0.70
     Raccoon
    0.69
     distressed
    0.68
     heartfelt
    0.68
    Activations Density 0.038%

    No Known Activations