© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13527
    Prev
    Next
    INDEX
    Explanations

    Mutation, Zahira

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

    code-related elements like file paths, stack traces, and programming errors in technical documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron fires on the “at” tokens that introduce lines in exception stack traces (e.g. Java stack‐trace frames).

    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
    
    -1.10
     следующем
    -1.10
     camisa
    -1.06
    %">
    -1.04
    %;">
    -1.03
     borboleta
    -0.97
     laissant
    -0.96
     ,
    -0.96
    ⫽
    -0.95
     lavable
    -0.94
    POSITIVE LOGITS
     and
    1.15
     unless
    0.96
    際に
    0.96
     estrenar
    0.96
     impl
    0.94
     at
    0.93
     Unless
    0.92
    باره
    0.91
     zetten
    0.90
     vermeld
    0.90
    Activations Density 0.006%

    No Known Activations