© 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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 16060
    Prev
    Next
    INDEX
    Explanations

    quantifiable measurements and units in scientific contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Levy
    -0.37
     baterías
    -0.35
     célé
    -0.35
     Pres
    -0.34
    ้ย
    -0.33
     تمت
    -0.33
     position
    -0.33
    rawDesc
    -0.33
     Plan
    -0.32
    ্দ
    -0.31
    POSITIVE LOGITS
     betweenstory
    0.63
     spoonful
    0.60
    posedge
    0.59
     otomatig
    0.58
    fuls
    0.58
    transQ
    0.56
     ***!
    0.56
     AssemblyTitle
    0.56
     للمعارف
    0.55
    ffions
    0.55
    Activations Density 0.515%

    No Known Activations