© 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. 10513
    Prev
    Next
    INDEX
    Explanations

    code from Nginx and related build scripts with a moderate focus on finding numbers

    oai_token-act-pair · gemini-2.0-flash

    digits

    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
    mag
    -0.55
    ']}
    -0.54
    ']:
    -0.53
    Leg
    -0.52
    Fram
    -0.50
    sim
    -0.49
    hat
    -0.49
    のだろうか
    -0.49
     Leg
    -0.49
    @[+][
    -0.48
    POSITIVE LOGITS
    RectangleBorder
    0.79
    ParallelGroup
    0.73
    VIAF
    0.66
     Мексичка
    0.65
    ContentAlignment
    0.64
    retudo
    0.63
    popd
    0.62
    GraphicsUnit
    0.58
    parametrize
    0.57
    MetaType
    0.57
    Activations Density 11.064%

    No Known Activations