© 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. 11600
    Prev
    Next
    INDEX
    Explanations

    light concepts

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

    This neuron flags infrequent, content‐rich words—especially weighty or thematic nouns (and sometimes verbs) that carry strong semantic meaning.

    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
    或是
    0.79
    もしくは
    0.68
    或者是
    0.67
     interdependent
    0.66
    lerinin
    0.65
    或
    0.65
     interrelated
    0.64
     hoặc
    0.64
     또는
    0.64
    individuals
    0.64
    POSITIVE LOGITS
    ใช้
    0.62
     glories
    0.62
     ગયા
    0.60
     antics
    0.60
     glory
    0.59
     kyun
    0.56
    inescence
    0.56
     герой
    0.55
     tutte
    0.55
     wares
    0.55
    Activations Density 0.000%

    No Known Activations