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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 15-GEMMASCOPE-RES-16K
    6. 1912
    Prev
    Next
    INDEX
    Explanations

    information related to announcements and inquiries about services or support

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     goof
    -0.33
     neglect
    -0.30
    hip
    -0.30
     sabia
    -0.29
    png
    -0.28
    ArrowToggle
    -0.28
     consider
    -0.28
    reiber
    -0.28
    有些
    -0.28
     forward
    -0.27
    POSITIVE LOGITS
    Personendaten
    0.88
     Wiktionnaire
    0.65
     estekak
    0.63
    󠁬
    0.62
    principalTable
    0.62
    ConstraintMaker
    0.62
    MigrationBuilder
    0.61
     Normdatei
    0.60
    istoitu
    0.59
    principalColumn
    0.58
    Activations Density 0.196%

    No Known Activations