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. 120814
    Prev
    Next
    INDEX
    Explanations

    the repeated phrase "of" in various contexts, indicating a focus on relationships or elements within categories or systems

    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
     faſt
    -0.60
     raiſ
    -0.54
     juſ
    -0.50
     ſta
    -0.50
     anſ
    -0.48
     houſe
    -0.47
     viſ
    -0.46
     ſtand
    -0.46
     visuel
    -0.46
     ſche
    -0.45
    POSITIVE LOGITS
     of
    1.12
     OF
    0.95
    of
    0.92
     của
    0.89
    ของ
    0.88
     Of
    0.86
    Of
    0.80
    KommentareTeilen
    0.79
    lof
    0.79
    オブ
    0.79
    Activations Density 1.365%

    No Known Activations