© 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. 20-GEMMASCOPE-RES-16K
    6. 5316
    Prev
    Next
    INDEX
    Explanations

    references to performance metrics and optimization issues in software

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

    technical performance measurements and metrics, especially related to computing (like FPS, shaders, benchmark numbers).

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered 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
     roues
    -0.30
     επικ
    -0.29
     informado
    -0.28
     températures
    -0.27
     pourtant
    -0.27
    iedene
    -0.26
     collègues
    -0.26
     roue
    -0.25
     erstmal
    -0.25
     tradiciones
    -0.25
    POSITIVE LOGITS
     CPU
    0.68
     latency
    0.66
    GOTREF
    0.63
    TryDecodeAsNil
    0.63
     bottlenecks
    0.63
    CPU
    0.61
     cpu
    0.61
    MemoryWarning
    0.60
     slowest
    0.60
     speeds
    0.59
    Activations Density 0.454%

    No Known Activations