© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15951
    Prev
    Next
    INDEX
    Explanations

    transform hierarchy

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

    The neuron activates on the literal token “transform,” spotting occurrences of that programming/configuration keyword.

    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
     The
    -2.19
    8
    -2.03
     expectation
    -2.02
    лі
    -1.96
    鯪
    -1.91
     этот
    -1.89
    有的
    -1.88
    </strong>
    -1.88
    有一
    -1.85
    宜しくお願い
    -1.84
    POSITIVE LOGITS
    </h5>
    2.42
    ｣
    2.30
    ácií
    2.28
    我们就
    2.23
     نیز
    2.20
    In
    2.17
     wrze
    2.13
    ة
    2.11
    リューション
    2.05
     обычно
    2.02
    Activations Density 0.013%

    No Known Activations