© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92356
    Prev
    Next
    INDEX
    Explanations

    journal citations with page ranges

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

    This neuron activates on numeric reference tokens (e.g. publication years, volume and page numbers) in citation entries.

    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
     some
    -1.63
     now
    -1.61
     as
    -1.60
     from
    -1.59
     Some
    -1.58
     Even
    -1.58
     what
    -1.55
     which
    -1.41
     in
    -1.39
    Some
    -1.37
    POSITIVE LOGITS
    ܆
    1.38
    푆
    1.31
     fluide
    1.22
     horrid
    1.21
     archite
    1.19
     accep
    1.17
    Adres
    1.16
     meub
    1.16
     apporter
    1.16
     inp
    1.16
    Activations Density 0.005%

    No Known Activations