© 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. 22-GEMMASCOPE-RES-131K
    7. 27092
    Prev
    Next
    INDEX
    Explanations

    code identifiers and proper nouns

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

    The neuron is detecting semicolon tokens (the “;” character) that terminate statements 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_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
     Probably
    -1.45
     superbly
    -1.44
     Perhaps
    -1.42
     unrelenting
    -1.42
     With
    -1.42
     Maybe
    -1.40
     expertly
    -1.39
     extraordinarily
    -1.38
     Often
    -1.38
     Although
    -1.38
    POSITIVE LOGITS
     verhind
    1.52
     gardes
    1.39
     trám
    1.37
     jezt
    1.34
    瘣
    1.31
     zoude
    1.28
     rác
    1.27
     erhi
    1.25
     GETS
    1.25
     viſ
    1.23
    Activations Density 0.006%

    No Known Activations