© 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. 10-GEMMASCOPE-RES-131K
    7. 44586
    Prev
    Next
    INDEX
    Explanations

    years in the 1940s

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

    The neuron strongly activates on four-digit year tokens (especially mid-20th-century dates) in historical contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    3
    -1.20
     as
    -1.16
     there
    -1.10
    2
    -1.03
     offerings
    -1.02
    Radius
    -1.02
    ০
    -1.01
     new
    -1.00
     notch
    -0.99
    _
    -0.97
    POSITIVE LOGITS
    着一个
    1.27
     temen
    1.27
     préserver
    1.25
    konos
    1.23
    eting
    1.21
    IFFS
    1.20
    rentina
    1.20
     meerdere
    1.18
    logError
    1.16
    ść
    1.14
    Activations Density 0.032%

    No Known Activations