© 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. 4202
    Prev
    Next
    INDEX
    Explanations

    fashion, trade, travel, style, plant, investor, soup, reform

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

    This neuron fires on the leading word of hyphenated compound descriptors (e.g. art-filled, food-grade, fish-killing, sea-gypsies).

    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_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
     *}\
    0.49
    ందం
    0.47
    ness
    0.46
     données
    0.46
     alternation
    0.45
    ítás
    0.44
     adjudication
    0.44
     pathophysiology
    0.44
     controllability
    0.44
     $\{
    0.43
    POSITIVE LOGITS
    oriented
    0.56
    用的
    0.56
    -
    0.55
    慜
    0.54
    ور
    0.51
    たっぷ
    0.51
    oled
    0.50
    écie
    0.50
    用の
    0.49
    amay
    0.49
    Activations Density 0.000%

    No Known Activations