© 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. 113014
    Prev
    Next
    INDEX
    Explanations

    Brockton Tomlin escort

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

    The neuron fires on rare or unusual subword fragments (uncommon vocabulary pieces), i.e. low‐frequency token substrings.

    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
     each
    -1.78
     use
    -1.33
    )
    -1.31
     with
    -1.27
     support
    -1.26
     yerine
    -1.26
    each
    -1.22
     improve
    -1.21
     instead
    -1.17
    Each
    -1.17
    POSITIVE LOGITS
    ssing
    1.34
     liger
    1.31
     baie
    1.28
    薷
    1.27
     cabine
    1.27
     broder
    1.24
     decidedly
    1.23
     quelques
    1.23
     tfsi
    1.23
     cuillère
    1.23
    Activations Density 0.144%

    No Known Activations