© 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. 12264
    Prev
    Next
    INDEX
    Explanations

    provides robustness and prevention

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

    The neuron fires on strongly promotional or persuasive words (e.g. want, start, actively, truly), i.e. language used to urge or entice the reader.

    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
    éntesis
    0.96
    ﻤ
    0.95
     ትንሽ
    0.86
    cześ
    0.84
    Pedido
    0.84
     Цуки
    0.83
    alisation
    0.82
    İK
    0.81
    ِين
    0.81
    𝗺
    0.81
    POSITIVE LOGITS
     insures
    1.09
     insuring
    0.91
     provides
    0.90
     preventative
    0.88
     insure
    0.84
     includes
    0.84
     contains
    0.82
     that
    0.80
     secures
    0.80
     which
    0.79
    Activations Density 0.004%

    No Known Activations