© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10393
    Prev
    Next
    INDEX
    Explanations

    possessive words and following context

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

    The neuron specializes in detecting the first-person plural possessive pronoun “our.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wech
    0.70
     лише
    0.68
     assh
    0.66
     bartenders
    0.65
     waiters
    0.62
     unserem
    0.62
    admins
    0.61
    ᔨ
    0.61
     ہماری
    0.61
     اہلِ
    0.60
    POSITIVE LOGITS
     recorded
    0.61
     accompanying
    0.58
     produced
    0.54
    EOF
    0.54
    ビデオ
    0.54
    recorded
    0.54
     put
    0.53
    produced
    0.52
    ISM
    0.52
    INCT
    0.51
    Activations Density 0.003%

    No Known Activations