© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 10896
    Prev
    Next
    INDEX
    Explanations

    scenario, idea, act

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

    This neuron responds to salient content nouns—especially abstract or topic-bearing nouns that carry the main ideas of a sentence.

    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
     ។
    0.77
     cuz
    0.76
     because
    0.75
     ಏಕೆಂದರೆ
    0.74
    ។
    0.74
     appunto
    0.71
     nhé
    0.70
     çünkü
    0.70
    ®.
    0.68
     ذریع
    0.68
    POSITIVE LOGITS
     cannot
    0.87
    cannot
    0.84
     musí
    0.75
    จะไม่
    0.73
     must
    0.71
     трябва
    0.69
     may
    0.68
     tuleb
    0.68
     MUST
    0.66
    MUST
    0.65
    Activations Density 0.337%

    No Known Activations