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. 2484
    Prev
    Next
    INDEX
    Explanations

    specific types of anti-inflammatory drugs and their roles in treatment

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

    NSAIDs and inflammation

    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
    StoryboardSegue
    -0.58
    XmlType
    -0.50
     đảo
    -0.46
    隔
    -0.45
    pperware
    -0.44
     Steck
    -0.43
     Duarte
    -0.43
    ographed
    -0.43
    üdis
    -0.43
    AutoresizingMask
    -0.43
    POSITIVE LOGITS
    profen
    0.57
     prostagland
    0.51
    lofen
    0.50
     ibuprofen
    0.49
     Inflammation
    0.46
    romyalgia
    0.45
     paink
    0.44
     aspirin
    0.43
     COX
    0.42
     NSA
    0.41
    Activations Density 0.037%

    No Known Activations