© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 3642
    Prev
    Next
    INDEX
    Explanations

    the two-letter sequence "sp"

    oai_token-act-pair · gemini-2.0-flash

    Sp

    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
     Wikimedijinoj
    -0.50
     size
    -0.49
    fileprivate
    -0.48
    Identyfik
    -0.48
    <bos>
    -0.48
    енно
    -0.47
     je
    -0.47
     Meksiku
    -0.44
     S
    -0.43
     V
    -0.43
    POSITIVE LOGITS
    følgelig
    0.70
    脚注の使い方
    0.68
     StatelessWidget
    0.65
     INTERESAR
    0.61
    ConstraintMaker
    0.57
    AnchorStyles
    0.54
     vedea
    0.54
     StatefulWidget
    0.53
    enumi
    0.51
     Efq
    0.51
    Activations Density 0.466%

    No Known Activations