© 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. 8257
    Prev
    Next
    INDEX
    Explanations

    lists, model, realism

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

    The neuron strongly responds to longer, content-heavy words—especially multi-syllable nouns or domain-specific terms.

    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
     их
    0.77
     è
    0.76
     their
    0.75
     їх
    0.75
     उनको
    0.75
     mereka
    0.71
     conséqu
    0.69
     அறிந்து
    0.69
     semua
    0.68
     знают
    0.68
    POSITIVE LOGITS
     pomocí
    0.75
    했
    0.71
    海岸
    0.69
     outfitted
    0.68
     kombin
    0.65
     방식으로
    0.64
    <b>
    0.63
    instruct
    0.63
     hjälp
    0.62
    썼
    0.62
    Activations Density 0.275%

    No Known Activations