© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69952
    Prev
    Next
    INDEX
    Explanations

    restores, returns, possibly, VZV

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

    Russian text, particularly words related to variable assignment, returns, or programming concepts in Cyrillic.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron detects Russian verbs formed with the “воз-/вос-” prefix.

    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
    B
    -2.55
    );
    -2.30
    )),
    -2.20
    揷
    -2.14
     när
    -2.14
     membag
    -2.14
    )}
    -2.13
    ”,
    -2.11
    ,’
    -2.11
    F
    -2.09
    POSITIVE LOGITS
     extravagance
    2.36
     числа
    2.28
    ケモン
    2.14
     Architekten
    2.14
    Hochspringen
    2.11
    would
    2.09
    鲠
    2.09
     θα
    2.08
    heria
    2.03
    ⬮
    2.03
    Activations Density 0.002%

    No Known Activations