© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13957
    Prev
    Next
    INDEX
    Explanations

    tv services

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

    The neuron fires on content‐bearing words (nouns, verbs, adjectives, etc.) rather than on grammatical or 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
     Atau
    0.81
    riendo
    0.80
    ⢿
    0.80
     otrzym
    0.78
    멎
    0.77
    rivi
    0.76
    జయ
    0.76
    Ꮺ
    0.76
     Maß
    0.75
    alámb
    0.75
    POSITIVE LOGITS
     bombarded
    0.70
     tảo
    0.65
    Torpedo
    0.65
    Essential
    0.64
    expanding
    0.63
     海
    0.62
     kaleidoscopic
    0.61
    历史上
    0.61
    噌
    0.61
     accret
    0.61
    Activations Density 0.000%

    No Known Activations