© 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. 66500
    Prev
    Next
    INDEX
    Explanations

    structural analysis or get your

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

    This neuron detects the presence of Twig template syntax (e.g. Twig tags and keywords like `{%`, `set`, `load`, `url`, `twig`).

    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
     jesu
    -1.47
     hunde
    -1.38
     ét
    -1.27
     corrom
    -1.26
     klim
    -1.25
     rocher
    -1.23
     soldat
    -1.23
     tersenyum
    -1.21
     minde
    -1.20
     vond
    -1.19
    POSITIVE LOGITS
     formerly
    1.29
    ult
    1.16
    carus
    1.12
     ehemal
    1.12
     главным
    1.10
     fuc
    1.09
     निम्
    1.09
     éste
    1.06
    有两
    1.05
    utdown
    1.02
    Activations Density 0.000%

    No Known Activations