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. 24-GEMMASCOPE-RES-16K
    6. 1040
    Prev
    Next
    INDEX
    Explanations

    references to large residential properties, specifically emphasizing features such as the number of bedrooms and overall size

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Luxury and exclusivity

    np_acts-logits-general · gemini-2.0-flash

    luxury guarded estates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cheaply
    -0.54
     cheap
    -0.52
     inexpensive
    -0.50
     popolare
    -0.43
     útiles
    -0.43
     lanterna
    -0.42
     handy
    -0.41
     popolari
    -0.41
     cheapest
    -0.40
     populaire
    -0.40
    POSITIVE LOGITS
     luxury
    0.82
    yacht
    0.81
     yacht
    0.80
    Luxury
    0.80
     luxurious
    0.80
     yachts
    0.80
     Luxury
    0.79
     caviar
    0.79
     Sotheby
    0.77
     luxus
    0.76
    Activations Density 0.465%

    No Known Activations