© 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. 22-GEMMASCOPE-RES-131K
    7. 64384
    Prev
    Next
    INDEX
    Explanations

    opposite states or change

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

    The neuron flags words describing directional or sign changes (e.g. more positive/negative, displacement direction).

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     mistura
    -1.07
     cliquer
    -1.07
     homicidio
    -1.06
    erobic
    -1.05
     absorción
    -1.03
     écrire
    -1.02
    معلومات
    -1.00
     ujarnya
    -1.00
     oferecer
    -1.00
     zdroj
    -1.00
    POSITIVE LOGITS
     opposite
    1.45
     vice
    1.15
     Conversely
    1.12
     противополо
    1.07
     conversely
    1.06
    nejšie
    1.02
     umgekehrt
    0.98
     reverse
    0.97
    Throwaway
    0.96
     when
    0.95
    Activations Density 0.029%

    No Known Activations