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-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 2327
    Prev
    Next
    INDEX
    Explanations

    references to PDF documents and related file formats

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Star
    -0.60
     Pl
    -0.59
     plating
    -0.57
     plat
    -0.57
     tố
    -0.57
    mber
    -0.56
     Tare
    -0.55
    Segoe
    -0.54
     Scal
    -0.54
     Sarko
    -0.54
    POSITIVE LOGITS
    pdf
    1.25
     PDFs
    1.14
     pdf
    1.10
    PDF
    1.06
     PDF
    1.04
    Pdf
    0.94
     Pdf
    0.85
    posób
    0.83
     Theſe
    0.82
     ainfi
    0.81
    Activations Density 0.105%

    No Known Activations