© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 6606
    Prev
    Next
    INDEX
    Explanations

    syntactical structures or patterns in object definitions and relationships in code

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

    Code or structured data

    np_acts-logits-general · gemini-2.0-flash

    structure delimiters

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     CreateTagHelper
    -0.94
    <unused68>
    -0.84
    <unused8>
    -0.84
    [@BOS@]
    -0.84
    <unused79>
    -0.84
    <unused28>
    -0.84
    <unused52>
    -0.84
    <unused41>
    -0.84
    <unused14>
    -0.84
    <unused16>
    -0.84
    POSITIVE LOGITS
    ，
    0.38
    、
    0.38
    top
    0.33
    while
    0.32
    impresa
    0.31
    P
    0.31
    H
    0.30
    b
    0.30
    binaan
    0.29
    besides
    0.28
    Activations Density 0.371%

    No Known Activations