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. 114894
    Prev
    Next
    INDEX
    Explanations

    specific measurements and metrics related to weight

    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
     betweenstory
    -0.94
    LookAnd
    -0.89
    providedIn
    -0.80
    脚注の使い方
    -0.80
    reportWebVitals
    -0.77
    TestingModule
    -0.73
    bootstrapcdn
    -0.73
    WaitGroup
    -0.71
    writeField
    -0.70
    
    -0.69
    POSITIVE LOGITS
     =",
    0.38
    EqualTo
    0.36
     <>",
    0.36
    NotEqualTo
    0.34
     égale
    0.26
    })`
    0.25
     directe
    0.23
     equals
    0.23
    XCTAssertEqual
    0.22
    cách
    0.22
    Activations Density 0.000%

    No Known Activations