© 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. 22-GEMMASCOPE-RES-131K
    7. 35020
    Prev
    Next
    INDEX
    Explanations

    else or otherwise conditions

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

    The neuron activates on numeric literals (tokens composed of digits) 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
    <>
    -1.01
     geograf
    -0.88
    घ
    -0.87
    ته
    -0.85
    StrictEqual
    -0.85
     Кара
    -0.85
     mêmes
    -0.84
    душка
    -0.84
     sirop
    -0.83
    СИ
    -0.83
    POSITIVE LOGITS
     otherwise
    2.30
     Otherwise
    1.76
     else
    1.68
    Otherwise
    1.58
     however
    1.51
     иначе
    1.42
    否则
    1.39
    else
    1.37
     other
    1.29
    otherwise
    1.28
    Activations Density 0.017%

    No Known Activations