© 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. 73525
    Prev
    Next
    INDEX
    Explanations

    historical eras and years

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

    The neuron activates for numeric year references, especially four-digit dates and decade markers.

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

    references to specific decades or time periods, particularly when formatted as "since the 1960s" or similar temporal phrases in historical contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    references to historical time periods—years, decades, centuries, and temporal qualifiers indicating when something occurred or since when it has occurred

    oai_token-act-pair · gpt-5Triggered 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
     Erkrank
    -1.48
    ⌀
    -1.46
     indicó
    -1.46
     витами
    -1.39
     erforder
    -1.38
     ранее
    -1.38
    最终
    -1.36
     überprü
    -1.34
     preſ
    -1.30
     detalló
    -1.28
    POSITIVE LOGITS
     or
    1.56
     after
    1.56
     both
    1.41
    ,
    1.35
    Pozri
    1.30
     two
    1.28
     began
    1.27
     碗
    1.23
     in
    1.21
     [
    1.21
    Activations Density 0.036%

    No Known Activations