© 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. 10-GEMMASCOPE-RES-131K
    4. 13934
    Prev
    Next
    INDEX
    Explanations

    artwork and biological structures

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

    This neuron detects the word "mosaic" and related terms, particularly in contexts of patterns, designs, or composite structures.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    ⡪
    -2.45
    sivité
    -2.36
    我們
    -2.22
     With
    -2.09
    ت
    -2.08
     она
    -2.08
     WITH
    -2.03
    ()]
    
    -2.03
    اگر
    -2.00
     wraz
    -2.00
    POSITIVE LOGITS
    and
    2.92
    '
    2.91
     of
    2.56
     formando
    2.50
     ferner
    2.33
     myriad
    2.22
     белом
    2.19
    接下来的
    2.17
     noin
    2.06
     zwölf
    2.00
    Activations Density 0.002%

    No Known Activations