© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 30-GEMMASCOPE-RES-16K
    6. 12688
    Prev
    Next
    INDEX
    Explanations

    opening and closing parentheses, as well as brackets

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

    Opening parenthesis in code

    np_acts-logits-general · gemini-2.0-flash

    programming keywords and identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     growth
    -0.47
     treatment
    -0.46
     erosion
    -0.45
     noqa
    -0.44
     rate
    -0.44
     row
    -0.43
     IndexError
    -0.43
     strife
    -0.42
    見ても
    -0.42
     tower
    -0.42
    POSITIVE LOGITS
    >>(
    0.57
    \_(
    0.56
    __(
    0.54
    )](
    0.54
    _(
    0.52
    ₁(
    0.52
     }^{(
    0.52
    ')(
    0.50
     ***!
    0.50
    (
    0.50
    Activations Density 0.613%

    No Known Activations