© 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. 65802
    Prev
    Next
    INDEX
    Explanations

    enclave code completion

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

    The neuron is primarily picking out programming‐or configuration‐style tokens—things like code identifiers, macro names, and include/file names.

    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
     nécessite
    -1.06
    odorant
    -0.99
    hassee
    -0.98
    alisis
    -0.94
    っていて
    -0.93
     巾
    -0.93
     moat
    -0.93
    yoruz
    -0.92
     devraient
    -0.90
     Эк
    -0.90
    POSITIVE LOGITS
     as
    0.92
     ●
    0.83
    ,’
    0.79
    `
    0.79
     معم
    0.77
     characters
    0.77
    =((
    0.77
    ules
    0.76
    ⋯
    0.75
     ''
    0.74
    Activations Density 0.580%

    No Known Activations