© 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. 36-GEMMASCOPE-RES-16K
    6. 15676
    Prev
    Next
    INDEX
    Explanations

    elements related to code execution and output formatting

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     estre
    -0.32
    glBind
    -0.32
     complesso
    -0.30
     paura
    -0.29
     muur
    -0.28
    就是这样
    -0.27
     üzere
    -0.26
     gray
    -0.26
     сбор
    -0.26
    kirakan
    -0.25
    POSITIVE LOGITS
     Administrativna
    0.73
     utafitiHapana
    0.66
    Panamoan
    0.61
    IndentedString
    0.59
    writeFieldEnd
    0.59
    RegistryLite
    0.58
     Italijanski
    0.57
     لينك
    0.57
    verwijspagina
    0.56
     MainAxisSize
    0.56
    Activations Density 0.178%

    No Known Activations