© 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. 33089
    Prev
    Next
    INDEX
    Explanations

    luscious masterful whimsical joyous wondrous

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

    The neuron strongly activates on long runs of leading whitespace (indentation) in code or comment blocks.

    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
     what
    -1.53
     out
    -1.43
     it
    -1.41
     their
    -1.36
    Jäh
    -1.35
    にほんブログ村
    -1.34
     to
    -1.34
     this
    -1.33
    When
    -1.30
     mildly
    -1.28
    POSITIVE LOGITS
    将她
    1.40
     luscious
    1.31
     masterful
    1.30
    将他
    1.23
     whimsical
    1.23
     tumultuous
    1.22
     Dinas
    1.20
     joyous
    1.17
     wondrous
    1.13
     sumptuous
    1.13
    Activations Density 0.001%

    No Known Activations