© 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. 10-GEMMASCOPE-RES-131K
    7. 105541
    Prev
    Next
    INDEX
    Explanations

    possessive their

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

    The neuron reliably fires on the plural possessive pronoun “their.”

    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
     two
    -1.70
    <eos>
    -1.48
     just
    -1.48
     "...
    -1.44
     "[
    -1.40
    ↵
    -1.39
     five
    -1.38
     make
    -1.36
     only
    -1.35
    ..."
    -1.35
    POSITIVE LOGITS
     themselves
    2.25
     are
    2.23
    themselves
    1.71
    secuted
    1.41
    疔
    1.41
    secta
    1.38
     besonderer
    1.36
    燹
    1.36
    ܞ
    1.33
     actuaciones
    1.32
    Activations Density 0.105%

    No Known Activations