© 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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13199
    Prev
    Next
    INDEX
    Explanations

    acknowledgments and references to contributors in scientific writing

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

    Text after commas and periods

    np_acts-logits-general · gemini-2.0-flash

    letters followed by period

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    MigrationBuilder
    -1.44
    Geplaatst
    -1.15
     noDo
    -1.13
    Personensuche
    -1.11
    abestanden
    -1.11
     ComVisible
    -1.07
     ModelExpression
    -1.07
    InjectAttribute
    -1.06
     transfieras
    -1.04
    ResumeLayout
    -1.03
    POSITIVE LOGITS
     J
    1.90
     M
    1.86
     A
    1.69
     R
    1.65
     C
    1.64
     E
    1.61
     P
    1.60
     L
    1.58
     S
    1.58
     D
    1.57
    Activations Density 0.855%

    No Known Activations