Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 9368
    Prev
    Next
    INDEX
    Explanations

    references to agents, particularly in the context of characters or roles in narratives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Krie
    -0.78
     BeautifulSoup
    -0.73
     bifur
    -0.70
     bifurcation
    -0.69
     battleship
    -0.68
     Cambo
    -0.67
     Lasso
    -0.67
     Leck
    -0.67
    prehension
    -0.67
    ths
    -0.67
    POSITIVE LOGITS
     agents
    1.94
     Agents
    1.87
     agent
    1.83
    Agents
    1.78
     Agent
    1.77
    Agent
    1.74
     AGENT
    1.68
     AGENTS
    1.66
    agents
    1.66
    AGENT
    1.59
    Activations Density 0.065%

    No Known Activations