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. 17-GEMMASCOPE-RES-65K
    6. 28545
    Prev
    Next
    INDEX
    Explanations

    colons and punctuation related to coding or data structure definitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yield
    -1.41
     yield
    -1.16
     yields
    -0.96
     Yield
    -0.96
    Yield
    -0.92
     Yields
    -0.90
     yielding
    -0.87
    yields
    -0.84
     yielded
    -0.68
    yielding
    -0.68
    POSITIVE LOGITS
    UserScript
    0.83
    ⓧ
    0.64
    ngdoc
    0.58
    
    0.52
    Personensuche
    0.52
    Osp
    0.49
     anyhow
    0.48
     anyway
    0.48
    anyway
    0.47
     tann
    0.46
    Activations Density 0.364%

    No Known Activations