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. 8580
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Dominick" and its variations, indicating a focus on a particular individual or brand associated with that name

    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
    fjspx
    -0.69
     surla
    -0.66
     '\\;'
    -0.66
    #+#
    -0.65
    ChildScrollView
    -0.57
     المعيارى
    -0.57
    msgTypes
    -0.57
     AttributeSet
    -0.56
     NSCoder
    -0.55
    RefNanny
    -0.54
    POSITIVE LOGITS
     DOM
    0.47
     dom
    0.46
    dom
    0.46
    DOM
    0.45
     Dom
    0.43
     domestically
    0.41
     domin
    0.39
     Domin
    0.39
    Dom
    0.38
     domes
    0.37
    Activations Density 0.299%

    No Known Activations