© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 14570
    Prev
    Next
    INDEX
    Explanations

    code, assembly language, and hardware descriptions

    oai_token-act-pair · gemini-2.0-flash

    Code and hardware configuration

    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
    SharedDtor
    -0.88
    InvalidProtocol
    -0.88
    Portail
    -0.87
    AndEndTag
    -0.87
    protoimpl
    -0.87
    tagHelperRunner
    -0.85
     oprot
    -0.84
    ChildScrollView
    -0.83
     <>",
    -0.80
    Autoritní
    -0.79
    POSITIVE LOGITS
     pret
    0.43
    Backup
    0.39
    nov
    0.38
    ٩
    0.38
    huis
    0.38
    Reg
    0.38
    Fre
    0.37
    とり
    0.37
    fre
    0.36
     Shift
    0.36
    Activations Density 0.936%

    No Known Activations