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. 73538
    Prev
    Next
    INDEX
    Explanations

    structural elements and syntax in programming or markup code

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

    Code syntax markers that denote structural elements in programming code, primarily structure delimiters like semicolons, brackets, and parentheses that complete definition syntax.

    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
    קישורים
    -0.50
     poveznice
    -0.45
    PCell
    -0.44
    IBOutlet
    -0.44
    }{*}{
    -0.43
     Wicidata
    -0.43
     estekak
    -0.42
    TextEditing
    -0.41
    躾
    -0.41
    WriteTagHelper
    -0.41
    POSITIVE LOGITS
    
    0.53
     Spitze
    0.45
    RTEX
    0.41
     zijne
    0.41
     Púb
    0.40
    StartTag
    0.39
    UnusedPrivate
    0.39
     atraves
    0.37
    pherals
    0.37
     hänen
    0.36
    Activations Density 0.060%

    No Known Activations