© 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. 13043
    Prev
    Next
    INDEX
    Explanations

    providing explanations

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

    This neuron highlights technical or specification-style nouns (e.g. “dimensions,” “graphics,” “control,” “storage,” “divider,” etc.) often found in product or procedural descriptions.

    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
    Mientras
    0.72
    RENDER
    0.63
    CONSTANT
    0.58
    EXPERIMENTAL
    0.58
     worthless
    0.56
    κέ
    0.56
    nonpayable
    0.55
    辟
    0.55
     Mientras
    0.54
    while
    0.53
    POSITIVE LOGITS
     proviso
    0.70
     agregado
    0.69
     embedded
    0.68
     وال
    0.67
     attached
    0.67
     Schwerpunkt
    0.66
     änd
    0.66
     ससुराल
    0.66
     incorporated
    0.65
     شرح
    0.63
    Activations Density 0.209%

    No Known Activations