© 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. 4-GEMMASCOPE-TRANSCODER-16K
    7. 4195
    Prev
    Next
    INDEX
    Explanations

    image file names with a specific naming convention

    oai_token-act-pair · gemini-2.0-flash

    5 or 7

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
    rétaire
    -0.55
     חיצוניים
    -0.54
    GOTREF
    -0.54
    setopt
    -0.54
     يتيمه
    -0.52
     Boletín
    -0.52
    abuhan
    -0.52
     Upstream
    -0.52
    \{\\
    -0.51
     ***!
    -0.51
    POSITIVE LOGITS
    IonicModule
    0.66
     bug
    0.56
     Bug
    0.54
    hoea
    0.52
    _'.$
    0.52
     Gew
    0.51
     bas
    0.49
     bình
    0.48
     Процитовано
    0.48
    Открыть
    0.47
    Activations Density 2.851%

    No Known Activations