© 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. 5147
    Prev
    Next
    INDEX
    Explanations

    This followed by a noun

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

    This neuron responds to technically dense, highly quantitative statements—especially those containing scientific terms or explicit numerical measures.

    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
     applauded
    0.80
    뤘
    0.80
     teşekkür
    0.79
    𝗹
    0.75
    𝙜
    0.75
    𝙤
    0.73
     journée
    0.73
     선보
    0.73
    𝗴
    0.72
     yaşanan
    0.72
    POSITIVE LOGITS
     Therefore
    0.91
     therefore
    0.88
    Therefore
    0.83
     Поэтому
    0.80
     поэтому
    0.79
     Thus
    0.79
     بنابراین
    0.73
     thus
    0.73
    このように
    0.73
    因此
    0.72
    Activations Density 0.567%

    No Known Activations