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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 40777
    Prev
    Next
    INDEX
    Explanations

    elements related to database operations and scripting

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

    Code syntax elements that connect or separate parts in programming languages, like special characters (colons in variable declarations, equals signs in assignments, brackets/braces in function definitions) and XML/HTML tags.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    PerformLayout
    -0.42
    ANTLR
    -0.36
    Derbyniad
    -0.34
     rf
    -0.34
    
    -0.34
    rich
    -0.32
    way
    -0.32
     verke
    -0.31
     AppBundle
    -0.31
    Aholisi
    -0.31
    POSITIVE LOGITS
    Personendaten
    0.64
     المعيارى
    0.57
     jsPsych
    0.57
     <<<<<<<<<<<<<<
    0.56
     propOrder
    0.55
    NameInMap
    0.50
    SharedCtor
    0.50
     again
    0.47
     الرياضيه
    0.47
    jsonString
    0.45
    Activations Density 0.383%

    No Known Activations