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-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 102998
    Prev
    Next
    INDEX
    Explanations

    HTML tags and attributes related to forms and structure within a web page

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     desmotivaciones
    -0.74
    tagHelperRunner
    -0.72
     pinulongan
    -0.69
    ScopeManager
    -0.69
    +#+#
    -0.63
     laſſen
    -0.63
    хьтан
    -0.62
    انجليز
    -0.61
    +#+
    -0.61
    SharedDtor
    -0.61
    POSITIVE LOGITS
    LayoutConstraint
    0.31
    fieldset
    0.31
    __()
    0.31
     verv
    0.30
    hogy
    0.28
    Form
    0.28
    )$/,
    0.28
    ))=\
    0.27
     lin
    0.27
    ായ
    0.27
    Activations Density 0.004%

    No Known Activations