© 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. 10055
    Prev
    Next
    INDEX
    Explanations

    difficulty or learning curve

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

    The neuron detects promotional or marketing language—positive, attention-grabbing adjectives and venue/product terms used in ad-style copy.

    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
     Yale
    0.80
     coales
    0.79
     Gasoline
    0.75
    깡
    0.75
     caffe
    0.75
     Chipotle
    0.75
     ud
    0.73
     Benz
    0.73
     shunt
    0.72
    Yale
    0.72
    POSITIVE LOGITS
    ور
    0.79
    🚻
    0.77
     শ্রীযুক্ত
    0.73
    दक्षिण
    0.71
     awaits
    0.70
     stärker
    0.70
     Annotated
    0.68
    ఏ
    0.68
    می
    0.68
    مه
    0.67
    Activations Density 0.001%

    No Known Activations