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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 900
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and their associated context

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

    punctuation marks used in questions and statements

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

    . ?

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     and
    -0.77
     
    -0.73
      
    -0.73
     (
    -0.71
    <strong>
    -0.69
     &
    -0.66
     or
    -0.65
     a
    -0.65
    <b>
    -0.63
     is
    -0.62
    POSITIVE LOGITS
    __':
    
    1.51
     مشين
    1.48
     autorytatywna
    1.44
     متعلقه
    1.37
     للمعارف
    1.36
    __":
    
    1.33
     فريبيس
    1.26
    rungsseite
    1.24
    aarrggbb
    1.23
     kasarigan
    1.23
    Activations Density 0.053%

    No Known Activations