© 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. 9637
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and symbols related to integrals and sums

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

    mathematical notation for integrals, particularly those with interval notation like [0,1] or limits like 0 to infinity.

    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
    WithIOException
    -0.52
    dafx
    -0.52
    StructEnd
    -0.50
     oprot
    -0.49
     MatDialog
    -0.48
     plazos
    -0.46
     ſon
    -0.46
    majánló
    -0.46
    iffion
    -0.43
    webtoken
    -0.43
    POSITIVE LOGITS
    ++)
    
    0.48
    TokenNameLPAREN
    0.42
    <thead>
    0.42
    __*/
    0.39
    __(/*!
    0.39
    ++){
    
    0.38
     отредактировал
    0.36
    ++)
    0.35
     --}}
    0.35
     propTypes
    0.34
    Activations Density 0.568%

    No Known Activations