© 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. 3101
    Prev
    Next
    INDEX
    Explanations

    scripting content and development

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

    The neuron activates on semantically rich content words—especially nouns, technical terms, and proper names—marking key topical terms in the text.

    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
    ;
    0.53
    ();
    0.52
    number
    0.50
    いい
    0.50
    ました
    0.49
    。
    0.48
    し
    0.48
    Number
    0.48
    எ
    0.48
    J
    0.48
    POSITIVE LOGITS
    特定の
    0.54
     victimes
    0.52
     différents
    0.52
     يصبح
    0.52
    以外の
    0.51
     특정
    0.50
     infestations
    0.50
     عمليات
    0.50
     svært
    0.50
    ాల్లో
    0.49
    Activations Density 0.298%

    No Known Activations