© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41577
    Prev
    Next
    INDEX
    Explanations

    Seismic shifts

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

    The neuron fires on isolated, high-salience “label” tokens—e.g. section-heading words or markup tag names.

    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
    řízen
    -0.90
     Furthermore
    -0.90
    驺
    -0.90
     Moreover
    -0.89
    ién
    -0.88
     jornalista
    -0.88
    ꙰
    -0.88
    vitational
    -0.88
    自体は
    -0.85
     herói
    -0.85
    POSITIVE LOGITS
     animal
    0.93
    日に
    0.91
    href
    0.90
    确定
    0.87
    ium
    0.87
    Instead
    0.87
     після
    0.84
    ̫
    0.81
    -
    0.81
    身体
    0.81
    Activations Density 0.000%

    No Known Activations