© 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-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 9685
    Prev
    Next
    INDEX
    Explanations

    the abbreviation "Sec" followed by a number, likely referring to legal or regulatory section markers

    oai_token-act-pair · gemini-2.0-flash

    Sec

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    Sinon
    -0.49
    ajoz
    -0.47
     mam
    -0.45
    στα
    -0.43
     «
    -0.42
     some
    -0.42
     Enter
    -0.42
     diep
    -0.42
    дат
    -0.42
     predecess
    -0.41
    POSITIVE LOGITS
    Personensuche
    0.88
    ThroughAttribute
    0.76
    IndentedString
    0.75
    RectangleBorder
    0.73
    RetentionPolicy
    0.72
     AssemblyCulture
    0.71
     TextInputType
    0.70
    AndEndTag
    0.69
    DebuggerNonUser
    0.68
    TagMode
    0.68
    Activations Density 0.583%

    No Known Activations