Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 67492
    Prev
    Next
    INDEX
    Explanations

    phrases and structures indicating ongoing actions or experiences related to performance and accountability

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

    number of students or programs

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    :✨
    -0.49
    WebElementEntity
    -0.48
     Савезне
    -0.47
     ویکی‌پدی
    -0.47
    GraphicsUnit
    -0.42
    Bibliograf
    -0.40
    EndGlobalSection
    -0.39
     samtidig
    -0.38
     Freien
    -0.35
     asemenea
    -0.35
    POSITIVE LOGITS
    RenderAtEndOf
    0.75
    [@BOS@]
    0.69
    <unused14>
    0.68
    <unused8>
    0.68
    <unused79>
    0.68
    <unused41>
    0.68
    <unused74>
    0.68
    <unused51>
    0.68
    <unused23>
    0.68
    <unused28>
    0.68
    Activations Density 0.637%

    No Known Activations