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. 73467
    Prev
    Next
    INDEX
    Explanations

    transitional phrases and connectors within the text

    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
    
    -0.60
    isPrime
    -0.52
     viewType
    -0.52
    ulongan
    -0.47
    Lü
    -0.44
     pii
    -0.43
    toCharArray
    -0.43
     assa
    -0.42
     acca
    -0.42
    createCanvas
    -0.42
    POSITIVE LOGITS
    tualmente
    0.54
    으며
    0.54
     있으며
    0.54
    omiast
    0.54
    하며
    0.49
     puissiez
    0.48
     البته
    0.47
    WaitGroup
    0.47
     deose
    0.47
    TestingModule
    0.45
    Activations Density 0.012%

    No Known Activations