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. 30519
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "scroll" and its variations in various contexts

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

    scrolling directions and UI elements

    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
    ey
    -0.51
     Nature
    -0.49
     Than
    -0.47
     Major
    -0.47
    ian
    -0.47
    نیم
    -0.45
     Nic
    -0.45
    ians
    -0.45
    Than
    -0.44
     Tigers
    -0.44
    POSITIVE LOGITS
    scroll
    1.14
     scroll
    1.09
    Scroll
    1.09
     Scroll
    1.04
    croll
    1.01
    crollView
    1.00
    SCROLL
    0.93
     scrolls
    0.90
    scrolling
    0.85
     scrolling
    0.83
    Activations Density 0.003%

    No Known Activations