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. 20-GEMMASCOPE-RES-16K
    6. 2995
    Prev
    Next
    INDEX
    Explanations

    phrases expressing gratitude and acknowledgment of assistance

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

    Acknowledgement or gratitude expressions

    np_acts-logits-general · gemini-2.0-flash

    expressions of thanks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     صوتيه
    -0.67
     autorytatywna
    -0.67
    ViewImports
    -0.66
    queryInterface
    -0.65
     beginnetje
    -0.62
    
    -0.61
    őle
    -0.60
    expandindo
    -0.60
    XmlAccessorType
    -0.59
     kasarigan
    -0.58
    POSITIVE LOGITS
     special
    1.23
    special
    1.11
     thank
    1.10
     thanks
    1.10
    Special
    1.06
     Special
    0.98
    Huge
    0.97
     want
    0.97
    big
    0.96
     spéciaux
    0.96
    Activations Density 0.177%

    No Known Activations