© 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. 5857
    Prev
    Next
    INDEX
    Explanations

    code snippets and links

    oai_token-act-pair · gemini-2.0-flash

    code and URLs

    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
     Reſ
    -1.20
    ſelf
    -1.19
     raiſ
    -1.17
     houſe
    -1.17
    ſelves
    -1.17
     ſche
    -1.13
     purpoſe
    -1.13
     deſt
    -1.10
     Majefty
    -1.09
     juſ
    -1.08
    POSITIVE LOGITS
     lenker
    0.55
     át
    0.46
     
    0.45
    Viited
    0.45
     ni
    0.43
     eher
    0.43
     сожалению
    0.42
     unter
    0.42
     me
    0.42
     trong
    0.42
    Activations Density 2.888%

    No Known Activations