© 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. 65732
    Prev
    Next
    INDEX
    Explanations

    start of phrases

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

    The neuron fires on comment and documentation keywords in code (e.g. “Created” and “Initialization”), i.e. words that mark sections or events in code comments.

    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
    檮
    -1.26
     wundersch
    -1.19
     tendência
    -1.18
     wiesz
    -1.17
     woda
    -1.13
     wirklich
    -1.13
    そろそろ
    -1.12
     dança
    -1.11
    AnchorStyles
    -1.11
     wła
    -1.11
    POSITIVE LOGITS
    と
    1.23
     olarak
    1.13
     ज्ञात
    1.13
    eamiento
    1.12
    起来的
    1.11
    Răsp
    1.09
     infantil
    1.08
     adulti
    1.05
    TRIBUN
    1.05
    思議
    1.05
    Activations Density 0.000%

    No Known Activations