© 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. 128156
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively detects indefinite articles ("a" and "an") in the text.

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

    a/an followed by descriptive word

    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
     of
    -2.38
     của
    -1.32
     vijf
    -1.30
     Jerusalén
    -1.22
     nabíd
    -1.22
     about
    -1.21
     pięk
    -1.21
     ofrece
    -1.19
     lanzó
    -1.17
     Turquía
    -1.17
    POSITIVE LOGITS
     profundas
    1.23
    命的
    1.08
     without
    1.02
    We
    0.94
     όπως
    0.94
    ↺
    0.94
    幽
    0.94
     cercana
    0.92
    tuigen
    0.91
    ingtones
    0.91
    Activations Density 0.036%

    No Known Activations