© 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. 14529
    Prev
    Next
    INDEX
    Explanations

    parrot mimicking language

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

    The neuron selectively activates on words naming forms or channels of communication (e.g. media, messages, propaganda, advertising, speaker, meetings).

    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
    т
    0.89
    0
    0.80
    rc
    0.77
    ست
    0.75
    а
    0.75
    кі
    0.75
    ехал
    0.75
    𝓾
    0.75
     茶
    0.74
    2
    0.73
    POSITIVE LOGITS
     slogans
    1.08
     propagand
    1.00
     propaganda
    0.99
     campaigns
    0.87
     persuas
    0.84
     banners
    0.83
    Campaign
    0.83
     propagating
    0.82
    ประชา
    0.81
     pungent
    0.81
    Activations Density 0.000%

    No Known Activations