© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 84802
    Prev
    Next
    INDEX
    Explanations

    specific types of fatty acids, particularly those that are unsaturated and long-chain

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

    fatty acids

    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
    setVerticalGroup
    -0.53
    Ори
    -0.47
    webElementXpaths
    -0.47
    awtextra
    -0.46
    toPromise
    -0.46
    astéro
    -0.45
    inst
    -0.45
    TintMode
    -0.44
    verifyException
    -0.43
     herum
    -0.43
    POSITIVE LOGITS
     fatty
    0.97
     Fatty
    0.81
    Fatty
    0.72
    stagland
    0.54
    oleic
    0.53
    boxylic
    0.50
    neurial
    0.48
    romyalgia
    0.48
     getAddress
    0.45
    ruptcy
    0.44
    Activations Density 0.017%

    No Known Activations