© 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. 7804
    Prev
    Next
    INDEX
    Explanations

    ruby class and method definitions

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

    This neuron activates on the indentation tokens (runs of whitespace) marking code block structure.

    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
    asiswa
    -1.72
     tetrachloride
    -1.70
     而
    -1.70
    н
    -1.66
     tillegg
    -1.63
     problemet
    -1.61
    ad
    -1.55
    あんまり
    -1.55
    },
    -1.55
    ihnachts
    -1.55
    POSITIVE LOGITS
    </em>
    2.47
    </h3>
    1.94
    </h2>
    1.73
     urme
    1.69
     used
    1.55
     Ajoutez
    1.52
     animado
    1.51
    ’
    1.50
    しました
    1.50
     affitto
    1.48
    Activations Density 0.034%

    No Known Activations