© 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. 14552
    Prev
    Next
    INDEX
    Explanations

    instances of articles (a, an, the) followed by nouns or adjectives, often within phrases expressing a relationship or condition

    oai_token-act-pair · gemini-2.0-flash

    a

    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
     most
    -0.73
    Most
    -0.71
    most
    -0.71
     any
    -0.70
     Most
    -0.63
     the
    -0.55
     colpa
    -0.52
     doigt
    -0.52
    MOST
    -0.52
     referenties
    -0.52
    POSITIVE LOGITS
     HasFactory
    0.82
     propOrder
    0.76
    IntoConstraints
    0.73
    awtextra
    0.72
     single
    0.68
    parsedMessage
    0.65
    üyada
    0.65
    :✨
    0.63
    Hentet
    0.63
    Datuak
    0.60
    Activations Density 2.975%

    No Known Activations