© 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. 38868
    Prev
    Next
    INDEX
    Explanations

    legal phrases and company suffixes

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

    The neuron responds to tokens that are uppercase (or non-Latin script) headings and metadata labels (e.g. section titles like “COURT” or foreign‐script words).

    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
     bahwa
    -2.70
     всіх
    -2.33
     что
    -2.33
    𝕄
    -2.20
    cipta
    -2.13
    entador
    -2.09
    缡
    -2.09
    ombang
    -2.08
    ließen
    -2.05
     &&
    -2.05
    POSITIVE LOGITS
    ING
    2.75
    t
    2.61
    at
    2.55
      
    2.42
    te
    2.36
    ne
    2.34
     -
    2.30
    per
    2.20
    to
    2.17
     
    2.17
    Activations Density 0.000%

    No Known Activations