Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 2518
    Prev
    Next
    INDEX
    Explanations

    keywords related to scientific research and medical conditions

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

    virus, Catholic, lesbian, injections, thyroid, pies, virus, receptor, widow, Facebook, knit

    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
    reportWebVitals
    -0.59
     ligiloj
    -0.48
    EndGlobalSection
    -0.45
     MainAxisSize
    -0.45
    printStackTrace
    -0.44
    transQ
    -0.44
     abstrait
    -0.43
    󠁿
    -0.43
    ]")]
    -0.43
     asiática
    -0.42
    POSITIVE LOGITS
     ब्रेकडाउन
    0.46
    engesch
    0.41
     impati
    0.40
    ViewImports
    0.38
    windowFixed
    0.38
    estacks
    0.36
     Bár
    0.35
    Hentet
    0.35
     fic
    0.35
     trom
    0.35
    Activations Density 0.128%

    No Known Activations