© 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. 16195
    Prev
    Next
    INDEX
    Explanations

    code constructs and technical terms

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

    The neuron strongly activates on technical or domain‐specific terms—especially programming identifiers, code symbols (like brackets or pointers), and mathematical vocabulary.

    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
    减肥
    -1.08
     confuso
    -1.05
     ს
    -1.02
     doloroso
    -1.00
     noix
    -0.98
     scris
    -0.98
    üyor
    -0.97
     pleno
    -0.96
     colspan
    -0.96
    esos
    -0.94
    POSITIVE LOGITS
     will
    1.23
     akhirnya
    1.08
     viendo
    1.03
     gummies
    0.96
     our
    0.91
    。
    0.91
     which
    0.90
    zyła
    0.90
     detalló
    0.89
     recibir
    0.89
    Activations Density 0.035%

    No Known Activations