© 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. 10760
    Prev
    Next
    INDEX
    Explanations

    terms relating to race and ethnic groups, especially 'black' and 'white'

    oai_token-act-pair · gemini-2.0-flash

    black

    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
    useEffect
    -0.44
     Sch
    -0.44
    PhysRevD
    -0.42
     ruines
    -0.41
     desenhos
    -0.41
     Arhivirano
    -0.40
     dorados
    -0.39
    modelBuilder
    -0.39
     största
    -0.39
    tagHelper
    -0.38
    POSITIVE LOGITS
    WriteTagHelper
    0.85
    IntoConstraints
    0.77
    PreferredItem
    0.73
    principalColumn
    0.66
    AnchorStyles
    0.65
    complexContent
    0.65
     חיצוניים
    0.65
    aarrggbb
    0.65
     StatefulWidget
    0.64
    FieldBuilder
    0.62
    Activations Density 0.835%

    No Known Activations