© 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. 6820
    Prev
    Next
    INDEX
    Explanations

    the key core concept

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

    The neuron fires on long, multi-syllable abstract or technical nouns (e.g. spreadsheet, compatibility, problem, solution).

    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
     ungkap
    0.64
    r
    0.59
    nej
    0.56
    🧖
    0.56
    .~
    0.55
     chuyên
    0.55
    ellate
    0.55
    igte
    0.54
    iw
    0.53
    <0x07>
    0.52
    POSITIVE LOGITS
     fallacy
    0.74
     diferenças
    0.73
     هذا
    0.71
     això
    0.68
     differences
    0.68
     cosas
    0.67
     wording
    0.66
     ceci
    0.66
    你是
    0.64
    副作用
    0.64
    Activations Density 0.207%

    No Known Activations