© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 5134
    Prev
    Next
    INDEX
    Explanations

    word count related patterns and structure in programming code

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

    Code or math snippets

    np_acts-logits-general · gemini-2.0-flash

    electricity and energy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/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.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     surla
    -0.67
    featureID
    -0.56
    EndGlobalSection
    -0.55
     ویکی‌پدی
    -0.53
    Hentet
    -0.52
    blond
    -0.52
    DockStyle
    -0.52
     HasFactory
    -0.52
    
    -0.51
    kegaard
    -0.50
    POSITIVE LOGITS
    Vegeu
    0.36
    latin
    0.34
    __':
    
    0.33
    然
    0.33
     sain
    0.33
    kär
    0.32
     Rear
    0.32
    재
    0.32
    onTap
    0.32
    migrationBuilder
    0.31
    Activations Density 0.168%

    No Known Activations