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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 9225
    Prev
    Next
    INDEX
    Explanations

    references to the term "Hook" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    ly
    -0.58
    ModelAdmin
    -0.54
    land
    -0.49
     doInBackground
    -0.49
    nocache
    -0.49
    och
    -0.48
    lijke
    -0.48
     BeautifulSoup
    -0.47
     ControllerBase
    -0.47
    Karriere
    -0.47
    POSITIVE LOGITS
    出版年
    0.80
     فريبيس
    0.74
    enderror
    0.71
    sterone
    0.68
    sizeCache
    0.67
    SizeF
    0.67
    تقاوى
    0.66
    ScopeManager
    0.65
     Scud
    0.65
     nahilalakip
    0.65
    Activations Density 0.445%

    No Known Activations