Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 8652
    Prev
    Next
    INDEX
    Explanations

    references to boat specifications and performance metrics

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

    boats and Mercury engines

    np_acts-logits-general · gemini-2.0-flash

    Mercury boat engines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    HomeAsUpEnabled
    -0.58
    ftagPool
    -0.55
    iecie
    -0.50
     Nightingale
    -0.50
    croll
    -0.49
     Bhu
    -0.49
    nyc
    -0.47
    AdapterFactory
    -0.46
    wikimedia
    -0.46
     Cymru
    -0.46
    POSITIVE LOGITS
     boat
    0.98
     lake
    0.88
    🚤
    0.83
    Boat
    0.80
    boat
    0.79
     boating
    0.79
     boats
    0.79
     Boat
    0.79
     Lake
    0.79
     ponto
    0.78
    Activations Density 0.116%

    No Known Activations