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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1071
    Prev
    Next
    INDEX
    Explanations

    coding concepts and syntax related to API interactions and data structures

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

    Code snippets interacting with APIs or servers

    np_acts-logits-general · gemini-2.0-flash

    API calls and server interactions

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
     BitmapFactory
    -0.32
     acompaña
    -0.26
    CloseOperation
    -0.26
     despedida
    -0.25
    lepiej
    -0.25
    🚭
    -0.24
     TRAN
    -0.24
    reactivex
    -0.24
    ngOnInit
    -0.23
     acompañar
    -0.23
    POSITIVE LOGITS
    脚注の使い方
    0.89
     nonUne
    0.84
    pushFollow
    0.81
    MigrationBuilder
    0.77
     للمعارف
    0.75
     estekak
    0.73
     <>",
    0.73
    EndGlobalSection
    0.72
    الحياه
    0.69
     تانيه
    0.68
    Activations Density 0.434%

    No Known Activations