© 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. 10-GEMMASCOPE-RES-131K
    7. 110751
    Prev
    Next
    INDEX
    Explanations

    related to Amount, Image, String, Pulse, Post

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

    The neuron fires on Swift type names—especially built-in generic collection types (Array, Dictionary) and their element types (Int, String, Any) in type annotations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Swift code that declares or initializes collection types—especially arrays and dictionaries—using bracketed type annotations, generic collection syntax, and empty collection literals.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    array type declarations in Swift code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    "
    -2.03
     \
    -1.84
    1
    -1.80
    $
    -1.70
     takich
    -1.70
     intrigu
    -1.63
    obtained
    -1.62
     unor
    -1.60
    ᕮ
    -1.58
    s
    -1.56
    POSITIVE LOGITS
     this
    2.30
     that
    1.93
     During
    1.65
    <td>
    1.64
     during
    1.64
    ementara
    1.61
     Schriftsteller
    1.55
     without
    1.53
     gives
    1.52
    瑧
    1.52
    Activations Density 0.005%

    No Known Activations