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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7406
    Prev
    Next
    INDEX
    Explanations

    SQL commands and statements related to database table structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fjspx
    -0.60
    awtextra
    -0.59
    IsMutable
    -0.54
     Administrativna
    -0.54
    windowFixed
    -0.54
    titleMargin
    -0.50
     offline
    -0.50
     CreateTagHelper
    -0.50
    ItemBackground
    -0.49
    Cordialement
    -0.48
    POSITIVE LOGITS
    migrationBuilder
    0.50
    HasIndex
    0.41
     ligiloj
    0.32
    wpdb
    0.31
     NIK
    0.30
     مواليد
    0.29
     محفوظة
    0.28
     typelib
    0.28
    principalColumn
    0.28
     stř
    0.27
    Activations Density 0.092%

    No Known Activations