© 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. 7826
    Prev
    Next
    INDEX
    Explanations

    words related to scientific studies and analysis, including results, differences, and potential factors

    oai_token-act-pair · gemini-2.0-flash

    terms related to scientific analysis and results

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

    reasons/possibilities

    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
     الحره
    -0.81
     AssemblyTitle
    -0.71
     NSCoder
    -0.71
    webElementGuid
    -0.71
    AddHtmlAttribute
    -0.67
    脚注の使い方
    -0.67
     виправивши
    -0.63
     StatefulWidget
    -0.61
     محفوظة
    -0.59
    HtmlAttribute
    -0.58
    POSITIVE LOGITS
     can
    0.65
     ref
    0.59
     Ref
    0.57
    bkz
    0.55
     gynhyrchwyd
    0.53
    MigrationBuilder
    0.53
     confirms
    0.52
    velocityY
    0.51
     bisa
    0.51
    tagHelperRunner
    0.51
    Activations Density 3.661%

    No Known Activations