© 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. 4318
    Prev
    Next
    INDEX
    Explanations

    Foundation models, Transformer architecture, development

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

    The neuron selectively activates on salient content words—especially domain-specific or topic-bearing nouns—rather than on common function words.

    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
     içerisinde
    0.93
     :).
    0.84
     bulunmaktadır
    0.83
     ০১
    0.82
    ပါသည်။
    0.79
     befindet
    0.78
    /
    0.77
     evidences
    0.77
     ณ
    0.74
     accordance
    0.74
    POSITIVE LOGITS
    使う
    0.94
     нужны
    0.84
     مثل
    0.84
    普及
    0.80
    を作る
    0.80
    人们
    0.79
     آمریکا
    0.79
     automakers
    0.78
    を使う
    0.76
    ennials
    0.75
    Activations Density 0.190%

    No Known Activations