© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 65050
    Prev
    Next
    INDEX
    Explanations

    citations and metadata

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

    The neuron fires on numeric tokens—especially publication years, volume/issue numbers, page counts, and other numeric metadata in references.

    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
    出来ます
    -1.18
    itude
    -1.13
     that
    -1.12
    thday
    -1.09
     and
    -1.09
     several
    -1.05
    gradable
    -1.00
     who
    -0.96
     agus
    -0.96
     both
    -0.95
    POSITIVE LOGITS
    鿬
    1.13
    helves
    1.11
     vermittelt
    1.09
     smallpox
    1.05
     [''],
    1.03
    越多
    1.02
     geprüft
    1.02
    ριθ
    1.01
    祤
    1.01
    autres
    1.00
    Activations Density 0.074%

    No Known Activations