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. 20-GEMMASCOPE-RES-65K
    6. 985
    Prev
    Next
    INDEX
    Explanations

    references to evidence and examination in legal contexts

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

    references to actions and examination related to sports and religious texts

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

    examine closely

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     propOrder
    -0.69
    oredCriteria
    -0.66
    GEBURTSDATUM
    -0.61
    первых
    -0.55
    észetes
    -0.54
    featureID
    -0.52
     onlyOwner
    -0.51
    XmlType
    -0.51
    :✨
    -0.51
     réussi
    -0.50
    POSITIVE LOGITS
     carefully
    0.80
    是否
    0.72
     exhaus
    0.69
     closely
    0.68
     thoroughly
    0.68
     adequacy
    0.67
     cuidadosamente
    0.67
    是否有
    0.66
     xem
    0.66
    caref
    0.65
    Activations Density 0.334%

    No Known Activations