Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 37-GEMMASCOPE-RES-16K
    6. 5789
    Prev
    Next
    INDEX
    Explanations

    terms related to resource allocation and efficiency

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DeleteBehavior
    -0.47
    /*
    -0.40
    #
    -0.40
    Климат
    -0.36
    вості
    -0.34
    Portály
    -0.33
    boutin
    -0.32
    setViewportView
    -0.32
    //
    -0.31
    gatan
    -0.31
    POSITIVE LOGITS
     resources
    1.91
    resources
    1.59
     effort
    1.53
     ressources
    1.50
     recursos
    1.43
     RESOURCES
    1.40
    Resources
    1.40
     Resources
    1.39
    资源
    1.32
    effort
    1.31
    Activations Density 0.546%

    No Known Activations