© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5079
    Prev
    Next
    INDEX
    Explanations

    attention all

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

    The neuron flags bits of text that read like health/diet advice—especially imperative or instructional sentences (e.g. “Always try to rotate the vegetables and meat…”).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     provident
    0.70
     bylaws
    0.67
     liquidation
    0.66
     ties
    0.66
     liquidated
    0.63
     tocc
    0.63
     tak
    0.63
     romper
    0.63
     tus
    0.61
    ++/
    0.61
    POSITIVE LOGITS
     Otherwise
    0.71
    󠁳
    0.70
     Essential
    0.68
     Surve
    0.64
    ubarb
    0.64
     した
    0.64
    Otherwise
    0.63
     Treatments
    0.62
    Easing
    0.62
     Jardin
    0.61
    Activations Density 0.001%

    No Known Activations