© 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-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 758
    Prev
    Next
    INDEX
    Explanations

    coding-related keywords and symbols

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

    programming-related statements or commands

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

    evidence,

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    avyzd
    -0.61
     hinweg
    -0.57
    principalColumn
    -0.57
     configureStore
    -0.56
    cydow
    -0.53
    Personendaten
    -0.52
     OGSÅ
    -0.52
     NgModule
    -0.51
     cuantas
    -0.51
    󠁢
    -0.50
    POSITIVE LOGITS
    fflush
    0.71
     trajets
    0.64
     CURIAM
    0.63
    setof
    0.61
    $__
    0.58
    Cyfeiriadau
    0.56
     AssemblyTitle
    0.56
     {},
    0.56
     الحره
    0.55
    derry
    0.55
    Activations Density 0.015%

    No Known Activations