© 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. 22-GEMMASCOPE-RES-131K
    7. 9710
    Prev
    Next
    INDEX
    Explanations

    variable assignment ` = `

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

    The neuron is spotting numeric literal constants (numbers) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     from
    -1.20
     berichten
    -1.13
    ########.
    -1.11
     our
    -1.10
    ^{}
    -1.01
    lankton
    -1.01
    velocityY
    -1.00
    ronpa
    -0.99
    [])
    -0.96
    </sub>
    -0.94
    POSITIVE LOGITS
    銬
    1.00
     récit
    1.00
    ほら
    0.95
     défi
    0.92
    zaam
    0.91
     astuces
    0.89
    Opin
    0.89
    Ils
    0.88
     recours
    0.87
     enthou
    0.87
    Activations Density 0.004%

    No Known Activations