Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 1259
    Prev
    Next
    INDEX
    Explanations

    elements related to customer service policies and product listings

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

    financial documents and policies related to applications

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

    document labels and forms

    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
    PreferredItem
    -0.67
    aarrggbb
    -0.63
    uride
    -0.59
    msgSender
    -0.56
    Erreferentziak
    -0.56
    Pyx
    -0.55
    tinyos
    -0.55
    pushFollow
    -0.52
    AddHtmlAttribute
    -0.51
    الإنجليزية
    -0.48
    POSITIVE LOGITS
     sheet
    0.96
     document
    0.94
     sheets
    0.91
     documents
    0.89
     booklet
    0.86
    heet
    0.81
     report
    0.76
     memorandum
    0.74
     documento
    0.74
    PerformLayout
    0.73
    Activations Density 0.245%

    No Known Activations