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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 98928
    Prev
    Next
    INDEX
    Explanations

    descriptions of reflective or shiny surfaces

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

    polished mirrored glass and windows

    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
    MessageTagHelper
    -0.59
    skog
    -0.48
    WaitGroup
    -0.48
     zac
    -0.45
     expended
    -0.44
     expend
    -0.43
    gnore
    -0.43
    ConstraintMaker
    -0.43
    inqui
    -0.42
     jScrollPane
    -0.42
    POSITIVE LOGITS
     glass
    0.50
     crystal
    0.48
    ImageContext
    0.47
     refractive
    0.45
     crystalline
    0.44
     cristal
    0.44
     glasses
    0.43
    glass
    0.43
    crystal
    0.42
     shatter
    0.41
    Activations Density 0.074%

    No Known Activations