© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111108
    Prev
    Next
    INDEX
    Explanations

    acts like a comparison

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

    the word "like" when used for comparison or similarity (as in "act like" or "runs like").

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron responds to uses of “like” following a verb to form a comparison or simile.

    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
    t
    -2.22
     time
    -1.88
     easier
    -1.80
     different
    -1.71
     inspiring
    -1.67
     out
    -1.66
     and
    -1.66
     experience
    -1.64
     immediate
    -1.62
     zahlreichen
    -1.60
    POSITIVE LOGITS
     a
    2.61
     normaux
    1.98
    ଃ
    1.77
     courants
    1.75
     voisins
    1.70
     semblables
    1.69
     fosfor
    1.69
    Faire
    1.69
     dezelve
    1.69
     rocha
    1.68
    Activations Density 0.024%

    No Known Activations