© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 3-GEMMASCOPE-RES-16K
    11. 5768
    Prev
    Next
    INDEX
    Explanations

    references to tricks or deceptive techniques

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SessionFactory
    -0.83
    
    -0.81
     Bleach
    -0.81
     Burrow
    -0.81
     crickets
    -0.78
     Bem
    -0.78
     Bourgoin
    -0.77
    MLLoader
    -0.77
    IContainer
    -0.75
     CRS
    -0.75
    POSITIVE LOGITS
     tricks
    1.43
     trick
    1.43
    trick
    1.32
    tricks
    1.29
     Trick
    1.17
     twist
    1.05
    Trick
    1.04
     truco
    0.99
    twist
    0.99
     twists
    0.98
    Activations Density 0.092%

    No Known Activations