© 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. 14226
    Prev
    Next
    INDEX
    Explanations

    connectors between phrases such as "and, also, but, thus, well as"

    oai_token-act-pair · gemini-2.0-flash

    academic exposition verbs

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

    explaining

    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
    IPA
    -0.57
     ویکی‌پدیا
    -0.52
    EDEFAULT
    -0.50
     utafiti
    -0.49
     Normdatei
    -0.48
    unsplash
    -0.48
    enedig
    -0.47
     Thebes
    -0.46
     dotyczy
    -0.46
    INSEE
    -0.45
    POSITIVE LOGITS
     presenting
    1.23
     explain
    1.10
     describe
    1.07
     tell
    1.03
     presentation
    1.02
     disclose
    1.02
     presented
    1.01
     show
    1.01
     explaining
    1.01
    provide
    1.00
    Activations Density 11.944%

    No Known Activations