© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 3-GEMMASCOPE-RES-16K
    11. 8330
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and operations involving variables and constants

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Grüsse
    -0.47
     jullie
    -0.38
     partij
    -0.37
     democrática
    -0.37
     współpra
    -0.35
     głó
    -0.35
    kowej
    -0.34
    kowym
    -0.33
     Decke
    -0.33
     comunista
    -0.33
    POSITIVE LOGITS
     betweenstory
    1.07
    OGND
    1.03
     zwiſchen
    1.01
    <unused52>
    1.01
    StoryboardSegue
    1.01
    𑄮
    1.01
    <unused14>
    1.01
    <unused21>
    1.00
    <unused17>
    1.00
    [@BOS@]
    1.00
    Activations Density 4.236%

    No Known Activations