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. 5742
    Prev
    Next
    INDEX
    Explanations

    technical terms related to software or system components

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

    user input parameters or code directives

    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
    <eos>
    -0.30
     joy
    -0.28
    -
    -0.28
    <b>
    -0.28
    öglichkeiten
    -0.27
    <i>
    -0.27
    قبال
    -0.26
    readAs
    -0.25
     fat
    -0.25
    __":
    -0.24
    POSITIVE LOGITS
    SBATCH
    5.31
    InjectAttribute
    0.98
     للاسماء
    0.94
    dafx
    0.93
    WithIOException
    0.89
    httphttps
    0.86
    WebServlet
    0.86
    ArgumentParser
    0.85
     nahilalakip
    0.85
    StructEnd
    0.81
    Activations Density 0.000%

    No Known Activations