© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 98522
    Prev
    Next
    INDEX
    Explanations

    code delimiters and structure

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

    The neuron strongly responds to punctuation and delimiter tokens—brackets, angle‐brackets, quotes, colons, etc.—in source 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
     invigor
    -0.92
     harmful
    -0.81
     scarcely
    -0.79
    ****
    -0.79
     profoundly
    -0.78
     алгорит
    -0.77
     noticeably
    -0.77
    時尚
    -0.76
     partially
    -0.76
     intensely
    -0.76
    POSITIVE LOGITS
    eventually
    0.98
    hii
    0.87
     EXPERIMENT
    0.86
    Loose
    0.86
    hnen
    0.85
    agonal
    0.84
    ⲓ
    0.83
    QRST
    0.83
    结束
    0.83
    impre
    0.83
    Activations Density 0.026%

    No Known Activations