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. 18-GEMMASCOPE-RES-16K
    6. 10554
    Prev
    Next
    INDEX
    Explanations

    synonyms and references related to web application frameworks and tasks in build automation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IsContent
    -0.73
    HideFlags
    -0.63
    tanleria
    -0.58
     StringTokenizer
    -0.54
     conci
    -0.52
    repaint
    -0.52
    })*/
    -0.52
    luence
    -0.52
    Spoljašnje
    -0.51
    enment
    -0.51
    POSITIVE LOGITS
    abestanden
    0.86
     otomatig
    0.67
     MaterialApp
    0.67
     wireType
    0.65
     listens
    0.53
    setVerticalGroup
    0.53
    urlpatterns
    0.52
    errHandler
    0.51
    ReusableCell
    0.51
     listen
    0.50
    Activations Density 0.021%

    No Known Activations