© 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. 24-GEMMASCOPE-2-RES-16K
    4. 6070
    Prev
    Next
    INDEX
    Explanations

    originates, narrative, meaning

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

    This neuron fires on isolated content words that begin a new line or paragraph (i.e. the first “real” word after a break).

    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
    께서
    0.92
    дного
    0.91
     CERN
    0.88
    т
    0.87
     cervix
    0.83
     GMC
    0.81
     Boynton
    0.80
     Fauci
    0.79
    ια
    0.78
     Loyola
    0.78
    POSITIVE LOGITS
    nosť
    0.79
    ↵
    0.79
     দুর
    0.67
    седнев
    0.64
     বাণিজ্য
    0.63
     član
    0.63
     درجة
    0.63
    už
    0.62
     பார்த்து
    0.62
    surf
    0.62
    Activations Density 0.000%

    No Known Activations