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

    Neuronpedia

    APICircuit 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. 5-GEMMASCOPE-RES-16K
    6. 2487
    Prev
    Next
    INDEX
    Explanations

    programming functions and methods related to user interface and graphical displays

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     /\.(
    -0.70
     nahilalakip
    -0.59
     وتسجيلات
    -0.59
    </thead>
    -0.57
    ніципа
    -0.56
     isComment
    -0.55
    awtextra
    -0.53
    jScrollPane
    -0.52
     Infórmanos
    -0.52
     secund
    -0.52
    POSITIVE LOGITS
    SequentialGroup
    1.31
    ContentAsync
    1.10
    PerformLayout
    1.04
    LayoutStyle
    1.00
    SuspendLayout
    1.00
    AutoSize
    0.97
    TagHelper
    0.93
    LabelTagHelper
    0.92
    MessageTagHelper
    0.92
    ViewFeatures
    0.86
    Activations Density 0.009%

    No Known Activations