© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 15381
    Prev
    Next
    INDEX
    Explanations

    The neuron intensely activates on content-bearing marketing or product–feature words (e.g. “gourmet,” “imported,” “airfares”) that serve as key descriptive keywords in promotional text.

    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
     большие
    0.76
     नसल्या
    0.73
    ὡ
    0.71
    ничный
    0.70
    isEnabled
    0.70
     промышленности
    0.70
     энерги
    0.69
    it
    0.69
     hydrogenation
    0.69
    two
    0.68
    POSITIVE LOGITS
    *{
    0.72
     pric
    0.62
     usc
    0.61
     [[[
    0.61
    }
    0.61
     clic
    0.61
    地元
    0.59
     combat
    0.58
    сної
    0.57
    問い
    0.57
    Activations Density 0.000%

    No Known Activations