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. 114605
    Prev
    Next
    INDEX
    Explanations

    programming-related initialization concepts

    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
    ]=>
    -0.49
     OFDb
    -0.47
    ftagPool
    -0.45
    awtextra
    -0.45
    tagHelperRunner
    -0.45
    resizingMask
    -0.44
    Personensuche
    -0.44
     OkHttpClient
    -0.43
    ClientRect
    -0.43
    aronder
    -0.42
    POSITIVE LOGITS
    INITIAL
    0.77
    Initial
    0.74
    setInitial
    0.73
     Initial
    0.73
    initial
    0.72
     initial
    0.71
     INITIAL
    0.71
    Intial
    0.64
     initiale
    0.61
    初始
    0.59
    Activations Density 0.045%

    No Known Activations