© 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. 2177
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to official documents, books, and reports, especially ones of record

    oai_token-act-pair · gemini-2.0-flash

    technical terms and programming language constructs

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

    creative works titles

    np_max-act-logits · gemini-2.0-flash

    titles

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    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
     itſelf
    -0.71
     Efq
    -0.63
    ſelves
    -0.55
    wiſe
    -0.54
     myſelf
    -0.54
     dieß
    -0.52
     doubtnut
    -0.52
     Reſ
    -0.51
    省市镇
    -0.50
     Anſ
    -0.50
    POSITIVE LOGITS
     of
    0.69
    AddTagHelper
    0.69
    bootstrapcdn
    0.67
    ("$.
    0.67
    RunAsync
    0.63
    ListItemIcon
    0.63
    addCriterion
    0.62
     '\\;'
    0.61
     negó
    0.60
    SpringBootTest
    0.59
    Activations Density 11.505%

    No Known Activations