© 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. 58463
    Prev
    Next
    INDEX
    Explanations

    decades like 1960s or 1970s

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

    The neuron detects the “s” suffix attached to years (e.g. in “1960s,” “1980s,” etc.), i.e. references to decades.

    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
    そういえば
    -2.52
    それにしても
    -2.30
     vinci
    -2.28
    詎
    -2.27
    雠
    -2.25
    というわけで
    -2.25
    <bos>
    -2.17
    ついでに
    -2.14
    さてさて
    -2.13
    蕯
    -2.13
    POSITIVE LOGITS
     impediment
    2.06
    新
    1.95
    、
    1.95
     '
    1.95
     belki
    1.91
    จะ
    1.91
    新加坡
    1.84
    f
    1.84
     altercation
    1.83
     arcane
    1.83
    Activations Density 0.014%

    No Known Activations