© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100042
    Prev
    Next
    INDEX
    Explanations

    figuring out

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

    The neuron strongly responds to occurrences of the phrasal verb “figure out” (and its morphological variants like “figuring out,” “figured out”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     zwei
    -1.74
     connus
    -1.72
     queste
    -1.64
     telles
    -1.59
     diesen
    -1.57
     hacen
    -1.55
     wenn
    -1.52
     две
    -1.52
     dando
    -1.51
     så
    -1.50
    POSITIVE LOGITS
     why
    2.03
     größere
    1.85
     how
    1.82
    end
    1.75
     that
    1.70
     whether
    1.66
     what
    1.64
    Also
    1.56
    ｿ
    1.53
    Some
    1.52
    Activations Density 0.054%

    No Known Activations