© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 22654
    Prev
    Next
    INDEX
    Explanations

    change the world

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

    The neuron is activated by various inflected forms of the word “change” (e.g. “change,” “changes,” “changed”).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    8
    -1.88
     a
    -1.87
     to
    -1.74
    7
    -1.73
    0
    -1.63
    6
    -1.63
     derzeit
    -1.60
    9
    -1.53
    4
    -1.47
    5
    -1.45
    POSITIVE LOGITS
     decorar
    1.73
     optique
    1.71
     decoración
    1.68
    紲
    1.65
     obce
    1.62
     aluminio
    1.56
     colorear
    1.55
    asties
    1.52
    蝙
    1.48
    了
    1.46
    Activations Density 0.038%

    No Known Activations