© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9768
    Prev
    Next
    INDEX
    Explanations

    ascending sub-sequences

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

    The neuron is broadly looking for “content” words—i.e. open‐class nouns, verbs, and adjectives carrying real meaning—rather than little 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
    猊
    0.86
    ised
    0.84
    waved
    0.84
    되었습니다
    0.82
    ized
    0.81
     Ça
    0.81
     Đình
    0.81
     sulfon
    0.80
     Bursa
    0.80
    ar
    0.79
    POSITIVE LOGITS
    がたくさん
    0.70
     colorChoice
    0.67
    ఆర్‌
    0.66
    ７
    0.66
    ૭
    0.65
    anyak
    0.64
    0.64
     exponencial
    0.62
    apping
    0.61
     [@
    0.61
    Activations Density 0.001%

    No Known Activations