© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 12828
    Prev
    Next
    INDEX
    Explanations

    phrases that include comparison and quantity using words such as "more", "or", "a", "even," and "un"

    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
     stessi
    -0.48
    さま
    -0.47
    ########.
    -0.47
    IUrlHelper
    -0.47
    endphp
    -0.47
    stdc
    -0.47
     adott
    -0.47
    fillStyle
    -0.46
     rör
    -0.46
    fjspx
    -0.46
    POSITIVE LOGITS
     AssemblyProduct
    0.74
     autorytatywna
    0.66
    Попис
    0.66
     AssemblyCompany
    0.58
    RectangleBorder
    0.58
     Học
    0.56
    artifactId
    0.53
    хьтан
    0.52
    getSeconds
    0.52
    forRoot
    0.52
    Activations Density 1.218%

    No Known Activations