© 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. 22-GEMMASCOPE-RES-131K
    7. 70663
    Prev
    Next
    INDEX
    Explanations

    references to documents

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

    This neuron responds to numeric reference tokens (e.g. patent or literature citation numbers and years).

    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
     by
    -1.00
    RESUMEN
    -0.99
    _{{\
    -0.91
    
    -0.90
    ):
    -0.87
    ',{
    -0.87
    ()).
    -0.86
    くんの
    -0.85
    *}[
    -0.85
    )=
    -0.84
    POSITIVE LOGITS
     mici
    0.99
     både
    0.90
    gelijke
    0.86
     regal
    0.85
     referenced
    0.84
     link
    0.82
     multe
    0.82
     병
    0.82
    рана
    0.82
     wiele
    0.79
    Activations Density 0.004%

    No Known Activations