© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 4336
    Prev
    Next
    INDEX
    Explanations

    location names and descriptive adjectives, perhaps within a travel context, and also notices the French article "la"

    oai_token-act-pair · gemini-2.0-flash

    La

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    NAME
    -0.47
    errorHandler
    -0.47
    gypti
    -0.46
    zeczy
    -0.46
    Namespace
    -0.44
    softc
    -0.43
    LETIN
    -0.43
    FillColor
    -0.42
    espé
    -0.42
    Personensuche
    -0.41
    POSITIVE LOGITS
    RegressionTest
    0.91
     cref
    0.70
     NSCoder
    0.69
    RenderAtEndOf
    0.68
    ########.
    0.66
    saraba
    0.66
     postIndex
    0.64
     StringTokenizer
    0.64
     typelib
    0.63
     يتيمه
    0.61
    Activations Density 0.132%

    No Known Activations