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. 91584
    Prev
    Next
    INDEX
    Explanations

    pieces of code related to file handling and system calls

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    <?
    -0.36
     bag
    -0.34
    IsContent
    -0.34
    5
    -0.33
     Not
    -0.33
     Bog
    -0.32
     Bad
    -0.32
     mark
    -0.32
     fat
    -0.32
     not
    -0.31
    POSITIVE LOGITS
    readline
    2.50
     readline
    1.63
    readLine
    0.96
    ReadLine
    0.90
    parsedMessage
    0.77
    nextLine
    0.70
    readFileSync
    0.66
    getline
    0.63
    はじめに
    0.63
    newline
    0.62
    Activations Density 0.002%

    No Known Activations