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. 53338
    Prev
    Next
    INDEX
    Explanations

    references to genetic interactions or overlaps in a scientific context

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

    greater than expected by chance

    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
     متعلقه
    -0.53
    口
    -0.49
     gouverneur
    -0.49
    PerformLayout
    -0.48
     poitrine
    -0.47
    Chron
    -0.46
    RRect
    -0.46
     bambina
    -0.46
    TryParse
    -0.45
     camicia
    -0.45
    POSITIVE LOGITS
    IntoConstraints
    0.70
     समीक्षाओं
    0.61
    fromnode
    0.61
     VizieR
    0.61
     InputDecoration
    0.60
    richment
    0.59
     distributions
    0.59
     statistic
    0.57
    hamshire
    0.57
    gehalt
    0.56
    Activations Density 0.965%

    No Known Activations