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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 12977
    Prev
    Next
    INDEX
    Explanations

    terms and structures related to programming and development, specifically in code snippets

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     Biôgrafia
    -0.55
     kaarangay
    -0.55
    EDEFAULT
    -0.52
    expandindo
    -0.49
     الرياضيه
    -0.49
    舺
    -0.47
     newOwner
    -0.47
     getPost
    -0.47
    脚注の使い方
    -0.45
     entgegen
    -0.45
    POSITIVE LOGITS
    ruecos
    0.43
    RetentionPolicy
    0.42
    Eloquent
    0.41
     financiero
    0.40
    RuleContext
    0.40
     EconPapers
    0.40
    ViewFeatures
    0.40
     tenue
    0.40
     StatelessWidget
    0.40
     negociaciones
    0.39
    Activations Density 0.307%

    No Known Activations