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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 75399
    Prev
    Next
    INDEX
    Explanations

    mathematical operations or expressions involving variables and numbers

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

    mathematical equations and expressions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     &___
    -0.48
    hape
    -0.38
     getItemCount
    -0.35
    BuilderFactory
    -0.34
    nius
    -0.34
    bral
    -0.33
     Tiny
    -0.33
     ElementRef
    -0.32
     trucos
    -0.32
    getExtras
    -0.32
    POSITIVE LOGITS
    TypedDataSet
    0.51
    tagHelperRunner
    0.50
     _)
    0.47
    StructEnd
    0.47
    kusen
    0.46
    sizeCache
    0.45
    hyrchwyd
    0.44
    hoeddwyd
    0.44
     realizacji
    0.42
    initializeApp
    0.42
    Activations Density 0.010%

    No Known Activations