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. 38-GEMMASCOPE-RES-16K
    6. 9743
    Prev
    Next
    INDEX
    Explanations

    references to "real" in various contexts, particularly related to time and data

    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
     Wikimedijinoj
    -0.54
     kaarangay
    -0.47
    SharedDtor
    -0.46
    TagMode
    -0.45
     really
    -0.45
    reid
    -0.44
    PreferredItem
    -0.43
    aarrggbb
    -0.43
    antMatchers
    -0.42
    DockStyle
    -0.42
    POSITIVE LOGITS
     estate
    0.94
    estate
    0.76
    ignment
    0.75
     Estate
    0.69
    izable
    0.69
    politik
    0.69
     ESTATE
    0.66
    location
    0.66
    igned
    0.65
    locating
    0.65
    Activations Density 0.141%

    No Known Activations