© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 15784
    Prev
    Next
    INDEX
    Explanations

    audit or rods

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

    This neuron activates on capitalized proper names or other named entities in the text.

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

    proper nouns, particularly surnames and personal names.

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

    tokens containing the letter sequence “od” or “ud,” especially within capitalized names and terms.

    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
    აბ
    0.65
    டனும்
    0.58
     trouverez
    0.54
    ম্ব
    0.54
     רו
    0.54
     கழக
    0.54
     Тому
    0.54
     Spinach
    0.53
    க்களை
    0.52
    ल्ल्या
    0.52
    POSITIVE LOGITS
    dy
    0.81
    yssey
    0.75
    ्ड
    0.72
    casts
    0.72
    gers
    0.71
    dery
    0.68
    Another
    0.65
    ný
    0.65
    rigues
    0.65
    ders
    0.64
    Activations Density 0.424%

    No Known Activations