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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 120341
    Prev
    Next
    INDEX
    Explanations

    specific references to legal or governmental proceedings and entities

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     PeEnEo
    -0.50
     IEnumerator
    -0.48
    
    -0.48
    uxxxx
    -0.47
    忑
    -0.46
     habet
    -0.46
    pectiva
    -0.46
    OrNil
    -0.45
     diría
    -0.44
    RectangleBorder
    -0.43
    POSITIVE LOGITS
    setVerticalGroup
    0.43
    SequentialGroup
    0.40
    ToAction
    0.39
    IntoConstraints
    0.38
     penta
    0.38
    Hozzáférés
    0.36
     estimés
    0.36
     propOrder
    0.36
    sizeCache
    0.36
     cons
    0.35
    Activations Density 0.125%

    No Known Activations