© 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. 22-GEMMASCOPE-RES-131K
    7. 26530
    Prev
    Next
    INDEX
    Explanations

    careers, OS, PCR, fashion, preschool, chlorophyll, tolerance, files, hydration

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

    terms associated with professional fields or disciplines.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron consistently lights up on relatively rare or domain‐specific “content” words—basically uncommon terms that stand out in the text.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     at
    -0.93
     by
    -0.87
     of
    -0.85
     back
    -0.84
    L
    -0.81
    r
    -0.80
    ts
    -0.79
     there
    -0.78
     all
    -0.78
    H
    -0.78
    POSITIVE LOGITS
    dokument
    0.98
     zás
    0.98
     corazon
    0.96
     étanche
    0.93
    ÜN
    0.92
     когато
    0.92
     serão
    0.90
     provincie
    0.89
    ilingual
    0.88
    conocimiento
    0.88
    Activations Density 0.092%

    No Known Activations