© 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. 31-GEMMASCOPE-RES-16K
    11. 11638
    Prev
    Next
    INDEX
    Explanations

    components related to physical motion and sensory experiences

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

    descriptions of things suspended or floating in the air.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    descriptions of things suspended or floating in the air.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    :✨
    -0.51
    艋
    -0.40
     recl
    -0.39
    рее
    -0.39
    LayoutPanel
    -0.39
    Toole
    -0.38
     Rakyat
    -0.38
    stagland
    -0.37
     Sire
    -0.36
     Export
    -0.36
    POSITIVE LOGITS
     air
    1.75
    air
    1.52
     airborne
    1.39
     powiet
    1.27
    Air
    1.25
     воздухе
    1.24
     vzdu
    1.24
    空气
    1.21
     lucht
    1.20
     Air
    1.20
    Activations Density 0.219%

    No Known Activations