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. 1228
    Prev
    Next
    INDEX
    Explanations

    code elements related to the Django framework and model definitions

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

    Python programming constructs, particularly related to class and model definitions

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

    authenticate user instance

    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
    NewUrlParser
    -0.71
     morire
    -0.63
    ็จ
    -0.58
    횟
    -0.56
     vastaan
    -0.55
    pointments
    -0.55
     نص
    -0.55
    LabelTagHelper
    -0.52
     frontale
    -0.51
    aarrggbb
    -0.51
    POSITIVE LOGITS
    Erstellt
    0.66
     ModelExpression
    0.63
    最快更新
    0.56
    mergeFrom
    0.55
     layui
    0.50
     Signalez
    0.50
    terday
    0.49
     Model
    0.48
    exels
    0.48
    تقاوى
    0.47
    Activations Density 0.019%

    No Known Activations