© 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. 22-GEMMASCOPE-RES-131K
    4. 68666
    Prev
    Next
    INDEX
    Explanations

    references containing years

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

    This neuron is tuned to pick out publication years (4-digit date numbers) in reference citations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     homicidio
    -1.26
     masculinos
    -1.04
     ,"
    -0.99
    uled
    -0.95
    upun
    -0.93
     usados
    -0.93
    ]=-
    -0.93
     bancaria
    -0.93
    france
    -0.93
     若
    -0.93
    POSITIVE LOGITS
    Географи
    1.11
    bình
    1.09
     these
    1.05
    cách
    1.04
    üte
    1.02
     those
    0.99
     chanteuse
    0.98
    üre
    0.98
    UEZ
    0.97
    ڿ
    0.97
    Activations Density 0.002%

    No Known Activations