© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11099
    Prev
    Next
    INDEX
    Explanations

    Don't expect

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

    The neuron strongly activates on the capitalized string “Don” (including instances as part of “Don’t”).

    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_31_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
    또한
    0.69
     ನೇ
    0.68
    ták
    0.67
    ますが
    0.66
    ます
    0.66
     또한
    0.63
    ום
    0.61
    
    0.60
     додат
    0.59
    แต่
    0.59
    POSITIVE LOGITS
    derdag
    0.88
    ahue
    0.79
    ovan
    0.76
    ateurs
    0.71
    ations
    0.70
    nelly
    0.70
    adee
    0.68
    ny
    0.67
    ated
    0.65
    atello
    0.64
    Activations Density 0.420%

    No Known Activations