© 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. 31-GEMMASCOPE-2-RES-16K
    7. 5328
    Prev
    Next
    INDEX
    Explanations

    numbered headings and bullet points

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

    This neuron is primarily detecting paragraph or line‐break boundaries (i.e. the blank/newline tokens marking the start of a new segment).

    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
     $>$
    0.59
     bề
    0.59
    ⎡
    0.59
     reng
    0.58
     $\}$
    0.57
    )^
    0.56
    })^
    0.55
    @(
    0.55
     siswa
    0.54
     aras
    0.54
    POSITIVE LOGITS
    CHAPTER
    0.69
    PART
    0.58
     seventeenth
    0.56
    Ἄ
    0.56
    Chapter
    0.55
    chapter
    0.55
     twentieth
    0.55
     CHAPTER
    0.55
     fourteenth
    0.55
     sixteenth
    0.54
    Activations Density 0.058%

    No Known Activations