© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9646
    Prev
    Next
    INDEX
    Explanations

    opening quotation marks

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

    The neuron strongly activates on the opening words of each new paragraph, i.e. the first token of a block of text.

    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
     None
    0.64
     The
    0.63
     Presumably
    0.63
     Local
    0.62
     I
    0.61
     Later
    0.61
     Neither
    0.60
     nbsp
    0.60
     (…)
    0.59
     Implicit
    0.59
    POSITIVE LOGITS
     "
    2.61
     “
    2.38
     "(
    2.34
     "[
    2.30
     “[
    2.28
     “(
    2.27
     "...
    2.16
     "¿
    2.14
    「
    2.09
     "'
    2.06
    Activations Density 0.450%

    No Known Activations