© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129333
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on mentions of “last year,” i.e. explicit references to the previous year.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    comparing year-over-year or period-over-period

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     from
    -1.59
    จาก
    -1.00
    に行きました
    -0.97
     culoare
    -0.96
    说起
    -0.96
     ventilación
    -0.95
    from
    -0.95
    ится
    -0.92
    に挑戦
    -0.92
    をご覧
    -0.91
    POSITIVE LOGITS
     Itali
    1.13
    તિ
    1.07
     laki
    1.06
    EXAMPLES
    1.03
    woners
    0.95
     Meksiko
    0.94
    rierten
    0.92
     previous
    0.92
     seksi
    0.90
    ′
    0.89
    Activations Density 0.015%

    No Known Activations