© 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. 32762
    Prev
    Next
    INDEX
    Explanations

    potential and possibility

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

    This neuron responds to the formal, fact-driven prose of a newswire‐style article (particularly a sports news report), activating strongly on words and punctuation typical of that register.

    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
     giu
    -1.38
    OOOOO
    -1.14
     varier
    -1.14
     aussit
    -1.11
     sentra
    -1.11
     miſ
    -1.09
    emphas
    -1.09
     RECOMM
    -1.08
     applau
    -1.07
     pochette
    -1.07
    POSITIVE LOGITS
     could
    2.45
     possibility
    2.39
     potential
    2.33
     potentially
    2.27
     might
    2.16
     possibilities
    2.06
     possible
    1.95
     may
    1.82
     возможно
    1.77
     would
    1.72
    Activations Density 0.180%

    No Known Activations