© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 1765
    Prev
    Next
    INDEX
    Explanations

    commas in close proximity to either numbers or words related to the passage of time

    oai_token-act-pair · gemini-2.0-flash

    gerunds or present participles

    np_token-act-pair-logits · gpt-4o-mini

    comma

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     greateſt
    -0.57
     casó
    -0.52
     enfans
    -0.51
    glEnd
    -0.50
     sahiptir
    -0.50
     itſelf
    -0.49
     gangen
    -0.49
     chiesto
    -0.48
    UpInside
    -0.48
     parezca
    -0.47
    POSITIVE LOGITS
     taking
    0.98
     making
    0.92
     producing
    0.90
     getting
    0.89
     being
    0.88
     setting
    0.88
     putting
    0.88
     using
    0.88
     creating
    0.88
     spending
    0.88
    Activations Density 1.288%

    No Known Activations