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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 45831
    Prev
    Next
    INDEX
    Explanations

    variable and function names commonly used in programming contexts

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

    Tree data structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    tagHelperRunner
    -0.79
     EconPapers
    -0.67
    Tikang
    -0.66
    Hentet
    -0.65
    
    -0.61
    adaptiveStyles
    -0.61
     utafitiHapana
    -0.61
    bootstrapcdn
    -0.58
     BitmapFactory
    -0.56
    VIAF
    -0.52
    POSITIVE LOGITS
     convencional
    0.51
     personalizadas
    0.49
     raisonnable
    0.49
     desconocido
    0.48
     olvidar
    0.48
     razonable
    0.48
     apropiado
    0.48
     oscuridad
    0.47
     individuales
    0.47
     doméstico
    0.47
    Activations Density 0.137%

    No Known Activations