© 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. 24-GEMMASCOPE-TRANSCODER-16K
    7. 2847
    Prev
    Next
    INDEX
    Explanations

    a mix of what looks like different code languages and mathematical notations

    oai_token-act-pair · gemini-2.0-flash

    Foreign language

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     in
    -0.54
     into
    -0.40
    Si
    -0.38
     wollen
    -0.37
    ly
    -0.35
     si
    -0.35
    In
    -0.34
    o
    -0.34
    vi
    -0.34
     In
    -0.34
    POSITIVE LOGITS
     Efq
    1.05
     Monfieur
    0.96
     myſelf
    0.96
     Jefus
    0.86
     démocr
    0.85
     pleaſure
    0.84
     Theſe
    0.83
     Inscrivez
    0.82
     itſelf
    0.81
     leaſt
    0.79
    Activations Density 2.500%

    No Known Activations