© 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. 129094
    Prev
    Next
    INDEX
    Explanations

    This neuron responds to structured metadata tokens—especially numbers, dates, commas, and other list‐ or infobox‐style elements in factual listings.

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

    recently or short stories describing past events

    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
    licante
    -1.02
     vidare
    -0.98
    şap
    -0.97
     hvilke
    -0.95
     ADDITIONAL
    -0.90
     både
    -0.88
     angusti
    -0.86
     Fauci
    -0.85
    ahhh
    -0.85
     tesla
    -0.84
    POSITIVE LOGITS
     recently
    1.35
    Recently
    1.23
     this
    1.05
    recently
    1.04
     recent
    1.04
     sponsored
    1.02
     courtesy
    1.00
     Recently
    0.98
     newly
    0.97
     recientemente
    0.97
    Activations Density 0.188%

    No Known Activations