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. 25-GEMMASCOPE-RES-16K
    6. 3934
    Prev
    Next
    INDEX
    Explanations

    instances of characters looking or gazing in a scene

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FunctionFlags
    -0.60
     estekak
    -0.60
     Akismet
    -0.53
    TagMode
    -0.52
    onAttach
    -0.50
     vrijwilli
    -0.49
     AssemblyCulture
    -0.49
    runApp
    -0.48
    ToAction
    -0.48
     "..\..\
    -0.48
    POSITIVE LOGITS
     intently
    0.73
     blank
    0.68
     lovingly
    0.66
     fixed
    0.64
     expect
    0.64
     upward
    0.63
     implor
    0.61
     upwards
    0.61
     down
    0.61
     fij
    0.61
    Activations Density 0.136%

    No Known Activations