© 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. 15535
    Prev
    Next
    INDEX
    Explanations

    self-referential and positive sentiment language, especially first person pronouns and words like perfect, learn, best, guess, positive, and good.

    oai_token-act-pair · gemini-2.0-flash

    non-English or specialized terms

    np_token-act-pair-logits · gpt-4o-mini
    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
     فريبيس
    -0.69
     GenerationType
    -0.61
    parsedMessage
    -0.57
    migrationBuilder
    -0.53
    queryInterface
    -0.50
     joaat
    -0.49
    richTextPanel
    -0.49
     Comercio
    -0.48
     Harmonic
    -0.46
    frameCount
    -0.46
    POSITIVE LOGITS
    awtextra
    0.71
    Tembelea
    0.57
    сутствие
    0.53
    omeness
    0.53
    клопе
    0.52
    /**
    0.51
    glColor
    0.51
     chap
    0.49
    UTELY
    0.49
    getFullYear
    0.48
    Activations Density 1.668%

    No Known Activations