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. 28507
    Prev
    Next
    INDEX
    Explanations

    paragraphs or sections beginning with a specific marker or tag

    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
    Kontrola
    -0.45
     vieta
    -0.45
    resto
    -0.43
    iteritems
    -0.43
    AxisAlignment
    -0.41
    UnknownFields
    -0.40
    ="#">
    -0.40
    Msp
    -0.40
    errorHandler
    -0.39
    sibili
    -0.39
    POSITIVE LOGITS
    出版年
    0.82
     الحره
    0.82
     تضيفلها
    0.82
     typelib
    0.81
     للاسماء
    0.81
    MigrationBuilder
    0.76
     дописавши
    0.75
    __':
    
    0.72
     تانيه
    0.71
     linkovi
    0.71
    Activations Density 0.015%

    No Known Activations