© 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. 72332
    Prev
    Next
    INDEX
    Explanations

    academic citations

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

    The neuron detects four-digit year tokens (publication dates) in academic citations.

    oai_token-act-pair · o4-miniTriggered 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
     that
    -1.96
     of
    -1.65
     it
    -1.39
     they
    -1.34
     who
    -1.33
     then
    -1.33
     since
    -1.32
     which
    -1.30
     would
    -1.30
     until
    -1.17
    POSITIVE LOGITS
    Kamu
    1.25
    porque
    1.23
    灃
    1.19
     naší
    1.16
     regardant
    1.15
    Jangan
    1.14
     nejen
    1.13
    Closest
    1.13
     nimmer
    1.10
    dentro
    1.09
    Activations Density 0.003%

    No Known Activations