© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6460
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively fires on direct second-person address—in other words, it flags occurrences of “you” and related phrasing that speak directly to 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
    尽管
    0.70
     sản
    0.64
     thậm
    0.64
     resulting
    0.63
     asla
    0.63
    ō
    0.61
     unscathed
    0.59
     achieved
    0.59
     consumables
    0.57
    4
    0.56
    POSITIVE LOGITS
     importants
    1.16
     wichtigen
    1.16
     önemli
    1.11
     wichtige
    1.09
     importante
    1.08
     مهم
    1.08
     belangrijke
    1.07
     важли
    1.06
     importanti
    1.05
     importantes
    1.04
    Activations Density 0.264%

    No Known Activations