© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13627
    Prev
    Next
    INDEX
    Explanations

    instructionact likeshould act aswill act asif you are abe sure to actwill pretend to bepretend you are a

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

    The neuron is triggered by email‐address components (e.g. the “@” symbol and surrounding address/domain tokens).

    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
    <eos>
    0.78
    or
    0.73
    。
    0.73
    app
    0.69
    Api
    0.69
    \|
    0.67
    api
    0.65
    2
    0.65
    card
    0.64
    ille
    0.63
    POSITIVE LOGITS
     diversos
    0.83
     dużo
    0.80
     verschied
    0.76
     resposta
    0.76
     cosidd
    0.75
     verschiedenen
    0.74
     மைய
    0.74
     diversas
    0.74
     terme
    0.74
     allerlei
    0.74
    Activations Density 0.016%

    No Known Activations