© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 14258
    Prev
    Next
    INDEX
    Explanations

    legal and bureaucratic jargon, especially legal code citations

    oai_token-act-pair · gemini-2.0-flash

    programming or technical phrases

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

    legal documents

    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
    Not in Any Lists

    No Comments

    Negative Logits
    migrationBuilder
    -0.54
    новништво
    -0.54
    sizeCache
    -0.53
    :✨
    -0.52
    ampiyon
    -0.51
    ExecuteReader
    -0.48
    bigliamento
    -0.48
     xlink
    -0.46
    TableAdapter
    -0.46
    mergeFrom
    -0.44
    POSITIVE LOGITS
    tvguidetime
    0.67
     useAppContext
    0.63
    لينكات
    0.60
    EndContext
    0.60
    SOUNDBITE
    0.53
     végétale
    0.52
    ]--;
    0.52
    corrhi
    0.52
    renderItem
    0.52
     bumi
    0.52
    Activations Density 2.360%

    No Known Activations