© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 12138
    Prev
    Next
    INDEX
    Explanations

    references to mathematical functions and operations related to complex analysis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Code or package names

    np_acts-logits-general · gemini-2.0-flash

    library calls or method access

    np_acts-logits-general · gemini-2.5-flash-lite

    instructions and setup text for LaTeX-formatted math problems, especially question prompts with placeholders and tables at the start of a problem statement.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <unused61>
    -0.20
     Internasional
    -0.19
     cuyo
    -0.18
     Budaya
    -0.18
    生
    -0.18
    V
    -0.18
     Turquía
    -0.18
    ↵↵
    -0.18
    <unused63>
    -0.17
    D
    -0.17
    POSITIVE LOGITS
    <unused74>
    1.88
    <unused41>
    1.88
    <unused14>
    1.88
    <unused8>
    1.88
    <unused52>
    1.88
    <unused79>
    1.88
    [@BOS@]
    1.88
    <unused23>
    1.87
    <unused16>
    1.87
    <unused3>
    1.87
    Activations Density 7.630%

    No Known Activations