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. 16-GEMMASCOPE-RES-16K
    6. 12130
    Prev
    Next
    INDEX
    Explanations

    positive and empowering expressions or endorsements

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

    Code, links, and file paths

    np_acts-logits-general · gemini-2.0-flash

    technical and programming terms

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     RoutedEventArgs
    -0.58
     оригіналу
    -0.55
    PreferredItem
    -0.53
     وتسجيلات
    -0.53
    sina
    -0.52
    ......”
    -0.51
    omitempty
    -0.50
    ="'.$
    -0.50
     Jokowi
    -0.50
    _'.$
    -0.49
    POSITIVE LOGITS
    webElementXpaths
    0.69
     Jefus
    0.64
    WireFormatLite
    0.62
     internetowa
    0.62
     Monfieur
    0.61
    shutil
    0.59
     juſt
    0.58
     članak
    0.55
     raiſ
    0.55
     CreateTagHelper
    0.54
    Activations Density 0.174%

    No Known Activations