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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53308
    Prev
    Next
    INDEX
    Explanations

    statistical results presented quantitatively

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

    numbers and percentages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     CreateTagHelper
    -0.76
    protoimpl
    -0.69
    httphttps
    -0.62
    Lähteet
    -0.58
     للاسماء
    -0.56
    TagMode
    -0.55
    حوالہ
    -0.54
    出版年
    -0.54
    outer
    -0.53
    GEBURTSDATUM
    -0.53
    POSITIVE LOGITS
    wości
    0.54
    bleib
    0.52
    ::~
    0.50
    nachron
    0.50
    ViewFeatures
    0.50
    vegas
    0.49
    AndFlush
    0.49
    Palmarès
    0.48
     ibunya
    0.46
    casio
    0.46
    Activations Density 0.059%

    No Known Activations