© 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. 3206
    Prev
    Next
    INDEX
    Explanations

    This neuron detects instances of the phrase “designed to” (i.e. identifying when something is described as being designed for a particular purpose).

    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
     ofthe
    0.68
     рома
    0.67
     П
    0.66
    edited
    0.65
     Lagi
    0.65
     Requested
    0.64
     of
    0.64
    dashed
    0.64
    пети
    0.63
     п
    0.61
    POSITIVE LOGITS
     fomentar
    0.93
     fördern
    0.89
     maximize
    0.88
     fosters
    0.86
     mục
    0.83
     maximise
    0.82
     umożliw
    0.82
    เพื่อให้
    0.81
     bertujuan
    0.81
     möglichst
    0.80
    Activations Density 0.126%

    No Known Activations