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. 18-GEMMASCOPE-RES-16K
    6. 1507
    Prev
    Next
    INDEX
    Explanations

    indications of health and safety measures, particularly related to mask-wearing and social distancing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    safety measures or distancing

    np_acts-logits-general · gemini-2.0-flash

    health and safety guidelines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tendre
    -0.58
    RenderAtEndOf
    -0.55
    XmlType
    -0.53
     revés
    -0.52
     geschlagen
    -0.50
     coû
    -0.50
     <>",
    -0.49
    enfance
    -0.49
    thood
    -0.48
     Siria
    -0.48
    POSITIVE LOGITS
     distancing
    1.39
     COVID
    1.35
     Covid
    1.31
    COVID
    1.23
    Covid
    1.23
     covid
    1.21
     social
    1.12
     distanci
    1.08
     socially
    1.08
     safety
    1.07
    Activations Density 0.039%

    No Known Activations