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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 7733
    Prev
    Next
    INDEX
    Explanations

    instances of historical events and entities related to governance and politics

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

    Titles (Bishop, Count, etc.) preceding names

    np_acts-logits-general · gemini-2.0-flash

    names and titles followed by surnames

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     Global
    -0.46
     Deep
    -0.42
     SIMBAD
    -0.42
     +
    -0.41
    disse
    -0.41
     global
    -0.41
    orig
    -0.41
    estial
    -0.41
    ấn
    -0.40
    Global
    -0.40
    POSITIVE LOGITS
    ########.
    0.96
     CreateTagHelper
    0.91
    ]")]
    0.79
    DockStyle
    0.78
    UserScript
    0.77
     cherchés
    0.77
    WriteTagHelper
    0.77
    SourceChecksum
    0.73
    writeFieldEnd
    0.72
    Geplaatst
    0.72
    Activations Density 0.207%

    No Known Activations