© 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. 10879
    Prev
    Next
    INDEX
    Explanations

    William followed by a name

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

    The neuron detects occurrences of the proper name “William,” i.e. it fires strongly on that author’s first name.

    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
    ૂર્ણ
    0.63
    attr
    0.56
    改进
    0.55
    InOut
    0.55
    libro
    0.54
    films
    0.54
    Filler
    0.54
    茉
    0.54
    filler
    0.53
    > </
    0.52
    POSITIVE LOGITS
    VIDENCE
    0.68
     deix
    0.59
     transducer
    0.59
     नेटवर्क
    0.56
     дро
    0.56
     ответить
    0.55
     bestimm
    0.55
     одним
    0.54
     stills
    0.54
     إليها
    0.54
    Activations Density 0.005%

    No Known Activations