© 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. 58352
    Prev
    Next
    INDEX
    Explanations

    nouns followed by code structures

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

    The neuron is primarily triggered by capitalized identifiers (class or model names) in code.

    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
    1
    -1.63
     not
    -1.51
    就有
    -1.44
    자인
    -1.38
    after
    -1.36
    Two
    -1.36
    5
    -1.35
     its
    -1.32
    7
    -1.31
    Just
    -1.30
    POSITIVE LOGITS
    を書いて
    1.39
    よりは
    1.32
     would
    1.30
    ceded
    1.30
    を書く
    1.30
    を書
    1.30
    simplify
    1.29
    MMMM
    1.27
    ⸰
    1.27
     expect
    1.26
    Activations Density 0.011%

    No Known Activations