© 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. 11378
    Prev
    Next
    INDEX
    Explanations

    introduces a clause

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

    The neuron strongly activates on proper names, dates, technical terms, and other named entities.

    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
    ”、
    0.82
    また
    0.62
     また
    0.61
     そんな
    0.61
    ங்களிலும்
    0.56
    ंद्रा
    0.55
     अशा
    0.54
    ôtel
    0.54
    そんな
    0.53
    ாலும்
    0.53
    POSITIVE LOGITS
     which
    5.75
    which
    5.59
     WHICH
    4.88
    Which
    4.69
     Which
    4.53
    ซึ่ง
    4.41
     которое
    4.28
     котором
    4.22
     которой
    4.19
     которых
    4.09
    Activations Density 1.904%

    No Known Activations