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. 92597
    Prev
    Next
    INDEX
    Explanations

    references to various types of "hubs" or central points of interaction

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

    hub and related words

    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
     Apri
    -0.41
     XXX
    -0.40
     Colorado
    -0.39
    ptian
    -0.37
    asthan
    -0.37
     Egyptian
    -0.37
    arsk
    -0.36
    Apri
    -0.36
    ctest
    -0.35
    />";
    -0.35
    POSITIVE LOGITS
     Hub
    1.43
     hub
    1.38
    Hub
    1.37
     HUB
    1.27
     hubs
    1.27
    hub
    1.23
    HUB
    1.17
     Hubbard
    0.98
     HubSpot
    0.85
    shub
    0.77
    Activations Density 0.002%

    No Known Activations