© 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. 72170
    Prev
    Next
    INDEX
    Explanations

    code syntax and variables

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

    The neuron strongly activates on rare or high-index vocabulary tokens—e.g. type-variable names, uppercase identifiers, numbers, and other infrequent words.

    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
     otro
    -1.00
     andere
    -0.94
     can
    -0.93
     acht
    -0.93
     parece
    -0.90
    altro
    -0.88
    aulted
    -0.86
     altre
    -0.85
    cVar
    -0.84
    Outros
    -0.84
    POSITIVE LOGITS
    continence
    1.14
     Ancona
    1.07
    EXHIB
    1.05
    uchten
    1.05
     bronz
    1.01
    鞴
    0.99
     قاعده
    0.99
    EMPLOY
    0.98
     DMX
    0.98
    Algeria
    0.96
    Activations Density 0.001%

    No Known Activations