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. 73483
    Prev
    Next
    INDEX
    Explanations

    instances of the word "regime" used in various contexts

    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
     Sparks
    -0.52
    Flowers
    -0.51
     Flowers
    -0.50
    Money
    -0.50
    Flower
    -0.49
    Water
    -0.48
    Andy
    -0.47
    white
    -0.47
    second
    -0.47
     Penny
    -0.47
    POSITIVE LOGITS
     regime
    1.34
     Regime
    1.30
     regimes
    1.09
    regime
    1.07
     régimen
    0.91
     regimen
    0.84
     régime
    0.77
     rež
    0.77
     reż
    0.72
     regimens
    0.68
    Activations Density 0.005%

    No Known Activations