© 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. 24-GEMMASCOPE-2-RES-16K
    7. 2150
    Prev
    Next
    INDEX
    Explanations

    listing things, names, or descriptions

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

    The neuron fires strongly on the very first content word of a new text segment or sentence.

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

    commas separating numbers in citations, references, or numerical data.

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

    directive or signpost language—imperatives, list numbers, and section labels that guide the reader’s attention.

    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_24_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
    e
    0.88
    ɖ
    0.87
    eel
    0.80
    eem
    0.77
     indetermin
    0.77
     erforder
    0.75
     indeterm
    0.74
    しておく
    0.74
     andere
    0.74
     erh
    0.74
    POSITIVE LOGITS
     міста
    0.81
    AMES
    0.77
     Fairview
    0.77
    𝗧
    0.76
    टिया
    0.74
    ція
    0.73
     nationals
    0.73
     devastation
    0.73
     Villages
    0.72
    тувати
    0.72
    Activations Density 0.000%

    No Known Activations