© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 86394
    Prev
    Next
    INDEX
    Explanations

    mini sandwich orbs

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

    The neuron detects plural noun forms by strongly activating on word stems plus their plural suffix (e.g. the “–s,” “–es,” or “–als” ending of a noun).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    -1.47
     today
    -1.42
    alahkan
    -1.34
     how
    -1.34
     articol
    -1.32
    W
    -1.32
    R
    -1.28
    notations
    -1.28
    дию
    -1.25
     acciaio
    -1.22
    POSITIVE LOGITS
     omdat
    1.58
    themselves
    1.45
     themselves
    1.44
     are
    1.43
     were
    1.40
    蜮
    1.32
    nissan
    1.29
    shutterstock
    1.23
    athread
    1.22
    beyaz
    1.21
    Activations Density 0.154%

    No Known Activations