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. 56749
    Prev
    Next
    INDEX
    Explanations

    punctuation and formatting-related elements in code

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

    code and punctuation

    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
     ujednoznacz
    -0.79
     ویکی‌آمباردا
    -0.76
    SBATCH
    -0.76
     AspNetCore
    -0.75
     Мексичка
    -0.74
     الحره
    -0.72
     AssemblyTitle
    -0.72
    parsedMessage
    -0.71
    rrggbb
    -0.69
    aarrggbb
    -0.69
    POSITIVE LOGITS
    <strong>
    0.39
    F
    0.32
     but
    0.31
    <h1>
    0.30
    <em>
    0.30
    *
    0.29
    CH
    0.29
    s
    0.29
    t
    0.29
    ロス
    0.29
    Activations Density 0.040%

    No Known Activations