© 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. 34-GEMMASCOPE-RES-131K
    7. 117379
    Prev
    Next
    INDEX
    Explanations

    historical decades

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

    The neuron selectively activates on numeric tokens—especially those representing years or dates.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    isfile
    -1.20
    があるので
    -1.08
    また
    -1.08
    いただけます
    -1.02
    お願い
    -0.97
    写真は
    -0.96
     totalité
    -0.95
    fié
    -0.94
    ！(
    -0.94
     indicó
    -0.93
    POSITIVE LOGITS
     late
    1.50
     early
    1.47
     twenties
    1.26
     tilgjenge
    1.16
     mid
    1.15
    osphere
    1.14
     era
    1.12
     and
    1.08
     latter
    1.06
    了很多
    1.05
    Activations Density 0.040%

    No Known Activations