© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70959
    Prev
    Next
    INDEX
    Explanations

    comparison words

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

    This neuron fires on the core words of analogical comparison phrases—especially the “X are to Y”/“X is to Y” construction (e.g. “are to the Oscars”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     also
    -0.93
    čkou
    -0.82
    filtr
    -0.81
    她们
    -0.81
    ();
    -0.79
     znajd
    -0.79
    ystems
    -0.76
    řád
    -0.76
     vì
    -0.75
    vám
    -0.75
    POSITIVE LOGITS
     like
    1.24
     resembles
    1.13
     akin
    1.09
     resemble
    1.06
    就像
    1.05
     reminiscent
    1.00
     likened
    0.99
     reminds
    0.97
     بالنسبة
    0.97
     analogous
    0.94
    Activations Density 0.033%

    No Known Activations