© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13037
    Prev
    Next
    INDEX
    Explanations

    reflexive pronouns

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

    The neuron responds strongly to common nouns in their plural form (i.e. words ending in an English plural “-s”).

    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
     দেখেছি
    0.61
    mdat
    0.52
    რობ
    0.52
    лк
    0.50
    }=(\
    0.49
    ɡ
    0.49
    ^{*}}{\
    0.49
     সর্বনাশ
    0.49
    ﺮ
    0.49
    ystycz
    0.48
    POSITIVE LOGITS
     itself
    4.94
    本身
    4.28
    本身的
    3.84
     themselves
    3.70
     himself
    3.42
     sendiri
    3.42
     نفسها
    3.42
    自身
    3.36
     자체가
    3.36
    自体
    3.34
    Activations Density 0.768%

    No Known Activations