© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5600
    Prev
    Next
    INDEX
    Explanations

    code and file paths, especially when involving periods

    oai_token-act-pair · gemini-2.0-flash

    code syntax elements and structures

    np_token-act-pair-logits · gpt-4o-mini

    code snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ouv
    -0.47
    SourceChecksum
    -0.45
     bocas
    -0.45
    Mediabestanden
    -0.43
    DeleteMapping
    -0.43
     माल
    -0.41
    ">//
    -0.41
     @"/
    -0.40
    StoreMessageInfo
    -0.40
    passes
    -0.39
    POSITIVE LOGITS
     مشين
    0.72
    AntiForgeryToken
    0.68
     AssemblyCulture
    0.65
    MigrationBuilder
    0.60
     CreateTagHelper
    0.59
    ToProps
    0.56
     yake
    0.55
    matchCondition
    0.55
    itespace
    0.55
    Vidite
    0.55
    Activations Density 0.535%

    No Known Activations