© 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. 129161
    Prev
    Next
    INDEX
    Explanations

    This neuron activates on parentheses in code, i.e. it detects “(” and “)” tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    specific keywords and errors

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     former
    -0.96
     altes
    -0.91
     identidade
    -0.91
     fabulous
    -0.90
     weißem
    -0.87
    nachron
    -0.87
    licates
    -0.87
     stunning
    -0.87
     picturesque
    -0.86
     victories
    -0.86
    POSITIVE LOGITS
    assertRaises
    1.04
     नहीं
    0.92
     warned
    0.85
     pozor
    0.84
     ราคา
    0.84
    guera
    0.82
     عدم
    0.82
    assertThrows
    0.82
     bicicletas
    0.79
     voelen
    0.79
    Activations Density 0.012%

    No Known Activations