© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41153
    Prev
    Next
    INDEX
    Explanations

    code keywords and directives

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

    This neuron fires on standalone period tokens—i.e. lines or tokens consisting of just “.” (often marking directives or separations 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
     The
    -2.36
     As
    -2.28
     So
    -2.28
    t
    -2.19
     On
    -2.19
     Both
    -2.17
     At
    -2.11
     is
    -2.05
     Now
    -2.03
     If
    -1.95
    POSITIVE LOGITS
    雠
    2.69
    臯
    2.48
    ſelves
    2.41
    粜
    2.38
    囪
    2.25
    𖤣
    2.23
    苼
    2.22
     ſy
    2.22
    阄
    2.20
    🧌
    2.20
    Activations Density 0.073%

    No Known Activations