© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 15229
    Prev
    Next
    INDEX
    Explanations

    programming syntax related to function definitions and template instantiation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    jgl
    -0.40
    usercontent
    -0.37
     EXISTS
    -0.37
     Otherwise
    -0.35
    Otherwise
    -0.35
     EXIST
    -0.34
     lenker
    -0.34
    fillType
    -0.34
    subpackage
    -0.33
     Rana
    -0.33
    POSITIVE LOGITS
     override
    0.52
    AndEndTag
    0.51
     throw
    0.51
    exitRule
    0.49
    httphttps
    0.49
     noexcept
    0.48
     require
    0.48
    EXCLUSIVE
    0.47
     EXCLUSIVE
    0.47
     PURE
    0.47
    Activations Density 0.363%

    No Known Activations