© 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. 22-GEMMASCOPE-RES-131K
    7. 66499
    Prev
    Next
    INDEX
    Explanations

    years and specific subjects

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

    This neuron activates on four-digit numeric tokens that look like years (e.g. “2002,” “2015,” “2019,” etc.).

    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
     four
    -1.39
     went
    -1.20
     three
    -1.17
     two
    -1.16
     fourteen
    -1.15
     only
    -1.14
     first
    -1.12
     vidas
    -1.12
    ÁT
    -1.08
    νομ
    -1.08
    POSITIVE LOGITS
     they
    1.07
    ↵↵↵
    1.06
    Especially
    1.02
    especially
    1.02
    包括
    1.00
     soprattutto
    0.99
    特に
    0.98
    ?
    0.97
     principalmente
    0.97
     especialmente
    0.96
    Activations Density 0.023%

    No Known Activations