© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 34-GEMMASCOPE-RES-16K
    11. 13987
    Prev
    Next
    INDEX
    Explanations

    structured programming elements and syntax components in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.34.hook_resid_post
    Hook Layer
    34
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     where
    -0.63
     où
    -0.48
     где
    -0.47
     gdzie
    -0.44
     ahol
    -0.42
     donde
    -0.42
     WHERE
    -0.39
     Where
    -0.39
     onde
    -0.38
     wherein
    -0.37
    POSITIVE LOGITS
    Personendaten
    0.67
    Aholisi
    0.59
     виправивши
    0.58
    ChildScrollView
    0.55
    featureID
    0.54
    ViewImports
    0.53
    IntoConstraints
    0.51
     MainAxisSize
    0.50
    afficheront
    0.50
     Italijanski
    0.50
    Activations Density 0.090%

    No Known Activations