© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73951
    Prev
    Next
    INDEX
    Explanations

    /** documentation markers

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

    The neuron fires on the opening of C‐style documentation comment blocks (i.e. lines beginning with “/**”).

    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
     as
    -1.56
     και
    -1.52
     have
    -1.41
    もある
    -1.40
     when
    -1.38
     by
    -1.37
     all
    -1.36
    立場
    -1.36
    方針
    -1.30
     with
    -1.29
    POSITIVE LOGITS
     maniere
    1.52
    方がいい
    1.46
    ほうがいい
    1.45
    おそ
    1.44
    ほうが
    1.42
    さぁ
    1.39
    明日は
    1.38
    あぁ
    1.37
    ようになります
    1.37
     manier
    1.34
    Activations Density 0.011%

    No Known Activations