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. 53262
    Prev
    Next
    INDEX
    Explanations

    references to web-related topics or technologies

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

    web service and science

    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
    ArgsConstructor
    -0.57
    testify
    -0.53
    idar
    -0.50
    BoxShadow
    -0.49
     CascadeType
    -0.48
    下载附件
    -0.47
    Compre
    -0.44
    DockStyle
    -0.44
    MockBean
    -0.44
    новить
    -0.43
    POSITIVE LOGITS
     lenker
    0.62
     AppColors
    0.61
     '\\;'
    0.59
     ostavi
    0.59
     кӀ
    0.58
    Diwedd
    0.58
    vician
    0.57
    évaluateur
    0.55
     كومونز
    0.54
    ùm
    0.54
    Activations Density 0.157%

    No Known Activations