© 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. 304
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on tokens that refer to people or individuals (often in formal roles or as stakeholders)—for example “patients,” “individual,” “your,” “their,” etc.

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

    possessive pronouns, particularly "your" and forms with apostrophe-s indicating possession.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    expressions of possession or ownership, especially genitive forms marked with apostrophes and possessive pronouns.

    oai_token-act-pair · gpt-5Triggered 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
     Đá
    0.33
    博士
    0.31
    2
    0.31
     Emphasis
    0.30
    美女
    0.30
    textrm
    0.29
    ussen
    0.29
     Verlag
    0.29
    ecta
    0.29
     También
    0.29
    POSITIVE LOGITS
     psyche
    0.54
     plight
    0.51
     misfortunes
    0.51
     predicament
    0.50
     doings
    0.49
     insecurities
    0.48
     sensibilities
    0.48
     wellbeing
    0.48
     vulnerabilities
    0.46
     individuality
    0.46
    Activations Density 1.204%

    No Known Activations