© 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. 20-GEMMASCOPE-RES-16K
    11. 1474
    Prev
    Next
    INDEX
    Explanations

    terms related to scientific and engineering concepts, particularly in the context of materials and processes

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

    technical terminology related to specialized systems or equipment (like airbags, scientific facilities, lighting, etc.).

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    technical terms within specialized domains like architecture, engineering, law, and government.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     itſelf
    -0.50
     aDecoder
    -0.49
     AssemblyTitle
    -0.48
     Bren
    -0.48
     betreft
    -0.47
     themſelves
    -0.47
    󠁣
    -0.47
    bunya
    -0.46
    hender
    -0.46
    brigens
    -0.46
    POSITIVE LOGITS
     bezeichneter
    0.39
     applications
    0.37
     proposals
    0.37
     versions
    0.34
     evaluations
    0.34
     designs
    0.34
     bağlı
    0.34
     staffers
    0.33
     treatments
    0.33
     sightings
    0.32
    Activations Density 1.200%

    No Known Activations