© 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. 34-GEMMASCOPE-RES-131K
    7. 77649
    Prev
    Next
    INDEX
    Explanations

    code section dividers

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

    The neuron fires strongly on code-structure directives and metadata (e.g. “using” statements, “#region” headers, section names like “Members”/“Methods”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     cintura
    -0.88
     postav
    -0.88
    にの
    -0.85
    υτό
    -0.85
     antica
    -0.84
     motiva
    -0.84
     domicili
    -0.84
     Zé
    -0.84
     squats
    -0.84
     civili
    -0.84
    POSITIVE LOGITS
     region
    1.16
    region
    1.12
    REGION
    1.09
    ↵↵
    1.06
     and
    1.04
    Region
    1.03
     région
    0.98
     section
    0.97
     Region
    0.95
     '</
    0.94
    Activations Density 0.029%

    No Known Activations