© 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. 1441
    Prev
    Next
    INDEX
    Explanations

    This neuron detects section headings or list‐item lines—i.e. standalone entries marked by bullets, numbers, or title-case/keyword prefixes.

    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
     カ
    0.71
     အ
    0.70
    IVERS
    0.69
     ネ
    0.69
     白
    0.68
     ペ
    0.67
     almac
    0.66
     細
    0.66
     ピンク
    0.66
    сных
    0.66
    POSITIVE LOGITS
    discard
    0.53
    ​​​​
    0.52
    போதும்
    0.49
    <unused63>
    0.48
    break
    0.47
    Second
    0.46
     ded
    0.46
     Second
    0.45
     fixed
    0.44
    ahead
    0.43
    Activations Density 0.088%

    No Known Activations