© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 30671
    Prev
    Next
    INDEX
    Explanations

    definitions introduced by where

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

    The neuron flags LaTeX‐style mathematical definitions and notation, especially the start of a “Let …” or “where …” clause and the surrounding math delimiters.

    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
    urlpatterns
    -1.03
     that
    -0.94
    encing
    -0.86
    sträger
    -0.84
    そちら
    -0.84
    hatenablog
    -0.82
    vikling
    -0.82
    strlen
    -0.81
    それが
    -0.81
    licitud
    -0.80
    POSITIVE LOGITS
     again
    1.06
     we
    1.02
    ここで
    0.98
    chene
    0.96
     novamente
    0.95
     here
    0.94
    这里的
    0.93
     الأخيرة
    0.91
     burada
    0.91
    Tentang
    0.91
    Activations Density 0.041%

    No Known Activations