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-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 184
    Prev
    Next
    INDEX
    Explanations

    terms related to saving time, money, or resources

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

    words or phrases related to saving or storing something.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    FunctionType
    -0.51
     ITU
    -0.50
    thor
    -0.49
    EntityType
    -0.47
     hurricanes
    -0.47
     userType
    -0.47
     Barton
    -0.46
     Orts
    -0.46
     Murdoch
    -0.46
    DialogInterface
    -0.46
    POSITIVE LOGITS
    save
    1.20
     save
    1.12
    Save
    1.09
     Save
    1.09
     SAVE
    1.08
    saving
    1.06
    saves
    1.05
    SAVE
    1.05
     saving
    1.02
    Saving
    1.01
    Activations Density 0.093%

    No Known Activations