© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 54359
    Prev
    Next
    INDEX
    Explanations

    legal citations

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

    This neuron detects four-digit numbers used as years or dates (e.g. “2009,” “1999,” “1986”) in legal citations and headings.

    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
    べて
    -0.79
    로그램
    -0.75
     VLAN
    -0.69
     پوش
    -0.68
     shippers
    -0.67
    leč
    -0.66
     TextInputType
    -0.66
    🦀
    -0.65
     dakota
    -0.65
    构造
    -0.65
    POSITIVE LOGITS
    ボタン
    0.75
     gắng
    0.74
    heroku
    0.71
     บาท
    0.69
    よかった
    0.67
    丐
    0.66
    riga
    0.66
    Эк
    0.64
    БУ
    0.63
     Kae
    0.63
    Activations Density 0.055%

    No Known Activations