© 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. 5623
    Prev
    Next
    INDEX
    Explanations

    rapidly evolving adaptability

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

    The neuron is highly sensitive to content‐bearing “named” or salient nouns (e.g. month names in date ranges, proper nouns, and other high-information words) rather than 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
    раи
    0.93
    ringen
    0.89
    een
    0.89
    бы
    0.86
     наиболее
    0.84
     шесть
    0.83
     hauptsächlich
    0.82
    น
    0.82
    ream
    0.81
    nine
    0.80
    POSITIVE LOGITS
     Ciò
    0.86
     Mentre
    0.73
     P
    0.72
     Questo
    0.70
     Acest
    0.69
    Alors
    0.68
     rendered
    0.67
     Quando
    0.67
     Mesmo
    0.65
     Questa
    0.64
    Activations Density 0.000%

    No Known Activations