© 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. 7920
    Prev
    Next
    INDEX
    Explanations

    Amanda Plummer, rendija experiment, send water

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

    The main thing this neuron does is spot personal names (capitalized first‐ and last‐name tokens).

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

    Scandinavian or non-English names, particularly those containing special characters or uncommon letter combinations.

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

    proper nouns, especially uncommon or diacritic-bearing personal names.

    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
    дий
    0.67
    đồng
    0.66
    нтен
    0.64
    ныя
    0.62
    សំ
    0.59
    дә
    0.58
    ذیر
    0.57
    dessä
    0.57
     Découvrez
    0.56
    дей
    0.55
    POSITIVE LOGITS
    ependence
    0.96
    ees
    0.89
    orsement
    0.73
    மிழ்
    0.70
    ele
    0.70
    URANCE
    0.69
    esign
    0.69
    epend
    0.68
    롱
    0.67
    orf
    0.66
    Activations Density 0.761%

    No Known Activations