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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1144
    Prev
    Next
    INDEX
    Explanations

    references to specific shark species

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

    terms related to marine biology, specifically focusing on dolphin behavior and sleep patterns

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

    marine mammals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     insects
    -0.56
     Biv
    -0.53
     Yorke
    -0.52
     Timur
    -0.51
     saj
    -0.50
    虫
    -0.50
     Insects
    -0.49
     έν
    -0.49
     insect
    -0.49
     Arms
    -0.48
    POSITIVE LOGITS
     whales
    1.12
     whale
    1.09
    Whale
    0.93
     dolphins
    0.93
     Whale
    0.89
     Whales
    0.89
     dolphin
    0.88
    whale
    0.86
    GEBURTSDATUM
    0.84
     Dolphin
    0.75
    Activations Density 0.079%

    No Known Activations