© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 38-GEMMASCOPE-RES-16K
    11. 8041
    Prev
    Next
    INDEX
    Explanations

    references to programming namespaces and libraries

    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
    ATURA
    -0.46
    cstring
    -0.45
     Wur
    -0.43
    rav
    -0.43
    itize
    -0.42
    Sc
    -0.40
    ëren
    -0.40
    ges
    -0.39
     Hank
    -0.39
    signatures
    -0.39
    POSITIVE LOGITS
    namespace
    0.61
     namespace
    0.57
    ThroughAttribute
    0.54
    RenderAtEndOf
    0.48
    setVerticalGroup
    0.48
    lccn
    0.47
    aarrggbb
    0.46
    UnsafeEnabled
    0.46
    ManagedBean
    0.46
     nakalista
    0.46
    Activations Density 0.497%

    No Known Activations