© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 2757
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on words used in privacy/confidentiality assurances or professional-service disclaimers (e.g. “worry,” “about,” “information,” “disclosing,” “only professional,” “reliable,” “contact”).

    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
     (
    0.65
    7
    0.64
     
    0.63
    9
    0.63
    2
    0.59
    1
    0.58
    6
    0.57
     C
    0.57
    8
    0.57
    -
    0.54
    POSITIVE LOGITS
    オプション
    0.82
    だけでなく
    0.72
     cashback
    0.72
    オンライン
    0.71
     тях
    0.70
    他們
    0.69
     Vestone
    0.68
     этих
    0.66
    だけで
    0.66
    த்தையும்
    0.66
    Activations Density 0.002%

    No Known Activations