© 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. 10-GEMMASCOPE-RES-131K
    7. 71391
    Prev
    Next
    INDEX
    Explanations

    X-ray

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

    The neuron fires on tokens beginning with “X-” followed by numerals (e.g. “X-214,” “X-190,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    (
    -2.88
    #
    -2.31
    {
    -2.23
     consisting
    -2.17
    which
    -2.05
    ),
    -2.03
     phun
    -1.99
    ”
    -1.99
    ſelves
    -1.98
    m
    -1.98
    POSITIVE LOGITS
    蓴
    2.25
     escuchadas
    2.08
    撵
    2.03
     penchant
    1.96
    ୌ
    1.95
     conocidas
    1.94
    ᠂
    1.91
     phony
    1.88
     تصریح
    1.88
     famosas
    1.87
    Activations Density 0.010%

    No Known Activations