© 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. 24-GEMMASCOPE-TRANSCODER-16K
    7. 5089
    Prev
    Next
    INDEX
    Explanations

    a variety of possible topics; there's no single topic that describes it

    oai_token-act-pair · gemini-2.0-flash

    say "only"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     only
    -2.44
    only
    -2.11
    Only
    -1.87
     Only
    -1.77
     ONLY
    -1.70
    ONLY
    -1.67
     только
    -1.55
     alleen
    -1.51
     tylko
    -1.48
     تنها
    -1.45
    POSITIVE LOGITS
     qu
    0.58
     que
    0.56
    -
    0.48
    Stokes
    0.46
    kun
    0.44
    Externé
    0.44
    apati
    0.42
     kun
    0.41
    multer
    0.40
    Cyfarwyddwr
    0.39
    Activations Density 2.336%

    No Known Activations