© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 20823
    Prev
    Next
    INDEX
    Explanations

    computational cosmology models

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

    The neuron strongly responds to specialized scientific and mathematical terminology—especially physics-specific jargon such as “models,” “scalar,” and “field.”

    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
     pēc
    -1.54
    ].
    -1.36
    ).
    -1.29
    喟
    -1.27
    𝑁
    -1.23
    ');
    
    -1.18
    :
    
    -1.18
     fabrik
    -1.16
     brug
    -1.16
    ultur
    -1.16
    POSITIVE LOGITS
     here
    1.36
    ctiva
    1.34
     we
    1.33
    我先
    1.31
     you
    1.30
    🪛
    1.30
    🧫
    1.30
    ريكي
    1.27
    érance
    1.26
     budeme
    1.26
    Activations Density 0.062%

    No Known Activations