© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 13398
    Prev
    Next
    INDEX
    Explanations

    language frequently used in legal documents relating to motions, appeals, trials, and court orders.

    oai_token-act-pair · gemini-2.0-flash

    to

    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
     catches
    -0.57
     vectorielles
    -0.56
     imprimée
    -0.55
    ASM
    -0.54
     parting
    -0.54
     culturelles
    -0.53
     hardening
    -0.52
     canadienne
    -0.52
     wraps
    -0.52
     européennes
    -0.52
    POSITIVE LOGITS
    Демографія
    0.67
    UrlResolution
    0.52
     NSCoder
    0.49
    romyalgia
    0.47
    DrawerLayout
    0.47
    OGND
    0.47
     كومونز
    0.46
    verständlich
    0.46
    tritte
    0.46
     ErrIntOverflow
    0.46
    Activations Density 1.737%

    No Known Activations