© 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. 20-GEMMASCOPE-RES-16K
    6. 10127
    Prev
    Next
    INDEX
    Explanations

    occurrences of mathematical expressions and operations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
    IntoConstraints
    -0.65
    msgTypes
    -0.65
    errHandler
    -0.60
    UrlResolution
    -0.55
    Билгалдахарш
    -0.53
     Chwiliwch
    -0.52
     hoạch
    -0.52
     informée
    -0.51
     IBOutlet
    -0.51
    ReusableCell
    -0.50
    POSITIVE LOGITS
     norvég
    0.37
     liderazgo
    0.31
     vectorielles
    0.31
     '{@
    0.30
    TintMode
    0.30
     виправивши
    0.30
    Rcpp
    0.28
     Siglo
    0.28
    Durham
    0.28
    jelasan
    0.28
    Activations Density 1.313%

    No Known Activations