© 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. 15218
    Prev
    Next
    INDEX
    Explanations

    source code, configuration snippets or build instructions

    oai_token-act-pair · gemini-2.0-flash

    programming and coding terms

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

    code library linking

    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
    eseorang
    -0.62
    qtype
    -0.59
    ResponseWriter
    -0.57
     חיצוניים
    -0.56
    ioen
    -0.54
    indale
    -0.53
    ebenarnya
    -0.49
    fortawesome
    -0.48
     }^{*}
    -0.48
    rouwen
    -0.47
    POSITIVE LOGITS
    DockStyle
    0.74
    postsleuth
    0.55
     tartalomajánló
    0.54
    AndEndTag
    0.53
    CppMethod
    0.53
    verifyException
    0.52
    WebElementEntity
    0.52
    WebServlet
    0.49
     leveled
    0.47
    tvguidetime
    0.47
    Activations Density 3.550%

    No Known Activations