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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 10296
    Prev
    Next
    INDEX
    Explanations

    code-like text with type declarations and variable names containing "fixed", "int", "uint", and "float"

    oai_token-act-pair · gemini-2.0-flash

    Data structure definitions

    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
     Pro
    -0.66
    Carriera
    -0.65
    AccessorTable
    -0.65
     To
    -0.64
    addGap
    -0.61
    QPalette
    -0.60
     Mill
    -0.60
     Ma
    -0.58
    RuleContext
    -0.58
     pro
    -0.58
    POSITIVE LOGITS
     foon
    0.63
     Eskimo
    0.56
     termica
    0.55
     itſelf
    0.55
     Stallion
    0.54
     potuto
    0.54
    vitis
    0.53
     étoient
    0.52
     raiſ
    0.51
     abstrait
    0.51
    Activations Density 1.878%

    No Known Activations