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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 15065
    Prev
    Next
    INDEX
    Explanations

    numerical values and their formatting in the context of dates or statistics

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

    Numbers or ordinal numbers

    np_acts-logits-general · gemini-2.0-flash

    sixth, 6, numbers, dates

    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
    utilisons
    -0.48
    ContentLoaded
    -0.48
     Monday
    -0.43
     newOwner
    -0.43
    Monday
    -0.42
    Lundi
    -0.42
     matel
    -0.41
     חז
    -0.41
    断
    -0.41
    alisasi
    -0.40
    POSITIVE LOGITS
     sixth
    0.75
    TestingModule
    0.73
     Friday
    0.72
    EndProject
    0.71
    第六
    0.69
     SIX
    0.68
     Six
    0.68
     Sixth
    0.67
     sexta
    0.67
    6
    0.67
    Activations Density 0.488%

    No Known Activations