© 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. 38738
    Prev
    Next
    INDEX
    Explanations

    describing things with a characteristic

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

    This neuron selectively activates on the word “like,” especially when used to indicate 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
    柃
    -2.30
    ﻥ
    -1.82
    urazione
    -1.81
    nąć
    -1.78
    鲮
    -1.74
    ଃ
    -1.67
     êtres
    -1.66
    Извест
    -1.65
    плу
    -1.64
    挲
    -1.63
    POSITIVE LOGITS
     experiences
    2.17
     wußte
    2.11
    8
    2.00
     publicidade
    1.95
     aislados
    1.95
     practices
    1.91
     erhi
    1.89
     qualities
    1.84
     achievements
    1.80
     spectacular
    1.79
    Activations Density 0.015%

    No Known Activations