© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4651
    Prev
    Next
    INDEX
    Explanations

    machine learning and data processing

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

    This neuron lights up on tokens naming concrete technical actions in a manufacturing or testing procedure (e.g. verbs like “fill,” “measure,” “compare,” “charged,” “protrude”).

    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
     فبراير
    0.85
     ווא
    0.84
    ビーチ
    0.82
    イラスト
    0.82
    復活
    0.82
     بج
    0.80
     consapevole
    0.80
    偲
    0.79
    verdad
    0.79
     Aussie
    0.78
    POSITIVE LOGITS
     seepage
    1.00
     hereinafter
    0.97
     inlaid
    0.94
     sintering
    0.92
     mononuclear
    0.92
     mesenteric
    0.91
     hemolysis
    0.91
     centrifugation
    0.91
     leukocyte
    0.89
     microorganism
    0.88
    Activations Density 0.009%

    No Known Activations