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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2943
    Prev
    Next
    INDEX
    Explanations

    technical terminology related to system configurations and software setups

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

    Linux distributions or installation instructions

    np_acts-logits-general · gemini-2.0-flash

    Linux operating systems

    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
    PerformLayout
    -0.92
    ArgsConstructor
    -0.70
    migrationBuilder
    -0.63
     autorytatywna
    -0.59
    허
    -0.58
    ForResult
    -0.58
    Билгалдахарш
    -0.56
     الرياضيه
    -0.54
    DockStyle
    -0.53
    InjectAttribute
    -0.52
    POSITIVE LOGITS
     Ubuntu
    0.82
     Linux
    0.68
     ubuntu
    0.68
    Ubuntu
    0.65
     Libre
    0.65
     Plasma
    0.65
     gnome
    0.64
     Debian
    0.64
     Fedora
    0.63
     Gnome
    0.62
    Activations Density 1.100%

    No Known Activations