© 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. 12500
    Prev
    Next
    INDEX
    Explanations

    legal or obligatory terms

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

    The neuron strongly activates on second-person addresses—particularly occurrences of the pronoun “you” (and related reader-addressing headers like “Information”).

    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
     litres
    0.68
     ,
    0.65
     +
    0.63
     եւ
    0.60
     quies
    0.59
     vapour
    0.59
     (`
    0.58
     litre
    0.57
     etc
    0.56
     ($(
    0.56
    POSITIVE LOGITS
     должны
    0.82
     مهمه
    0.80
     должна
    0.79
     abogado
    0.78
     책임
    0.77
     повинні
    0.77
     повин
    0.76
     трябва
    0.75
     responsabilidad
    0.75
     attorney
    0.73
    Activations Density 0.000%

    No Known Activations