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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 13798
    Prev
    Next
    INDEX
    Explanations

    statements about international relations and political actions involving Russia and other countries

    oai_token-act-pair · gemini-2.0-flash

    capital cities

    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
    AndEndTag
    -0.69
    ]")]
    -0.68
     uParam
    -0.60
    ♥♥
    -0.55
     GenerationType
    -0.53
     sī
    -0.49
    IBinder
    -0.49
     AssemblyProduct
    -0.49
    isNameExpr
    -0.48
    ónde
    -0.48
    POSITIVE LOGITS
     Moscow
    1.40
     Ottawa
    1.27
     Beijing
    1.27
     Washington
    1.27
     Islamabad
    1.22
     Canberra
    1.15
     Tehran
    1.13
     Jakarta
    1.13
     Berlin
    1.12
     Ankara
    1.12
    Activations Density 2.965%

    No Known Activations