© 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. 98361
    Prev
    Next
    INDEX
    Explanations

    programming language keywords

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

    The neuron is looking for single‐letter variable identifiers (and their declarations/uses) 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
     yurt
    -1.10
    nisk
    -1.08
    😶
    -1.03
     abusi
    -1.02
    ävän
    -1.02
    ᴩ
    -1.01
    侧面
    -1.01
    abuhan
    -1.00
    潸
    -1.00
     posando
    -0.99
    POSITIVE LOGITS
     solution
    1.33
     solutions
    1.28
     even
    1.22
     also
    1.13
     two
    1.10
    ższych
    1.09
     Game
    1.09
     game
    1.09
     I
    1.08
     for
    1.04
    Activations Density 0.014%

    No Known Activations