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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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 - 16k
    5. 10-GEMMASCOPE-RES-16K
    6. 2778
    Prev
    Next
    INDEX
    Explanations

    references to "dark" and "dark matter," indicating a focus on themes related to darkness and its impacts or representations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AssemblyTitle
    -0.98
     propOrder
    -0.81
    ↩
    -0.80
    tyfik
    -0.78
    vedra
    -0.74
    mfrac
    -0.74
    osoever
    -0.74
    Gilla
    -0.73
     protoimpl
    -0.73
     Schlegel
    -0.72
    POSITIVE LOGITS
     Dark
    1.53
     DARK
    1.51
    DARK
    1.42
    Dark
    1.42
     dark
    1.37
    dark
    1.28
    Darkness
    1.21
     darkness
    1.17
    darkness
    1.12
     Darkness
    1.11
    Activations Density 0.051%

    No Known Activations