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. 18-GEMMASCOPE-RES-16K
    6. 15145
    Prev
    Next
    INDEX
    Explanations

    code and programming-related terminology

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

    Code snippets defining data structures/tables/charts

    np_acts-logits-general · gemini-2.0-flash

    currency and index mappings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PositiveButton
    -0.49
    нгредіє
    -0.47
    رشف
    -0.45
     Тек
    -0.43
    toral
    -0.42
     doInBackground
    -0.42
    conci
    -0.40
    unteer
    -0.40
    NewGuid
    -0.39
    msgSender
    -0.38
    POSITIVE LOGITS
     table
    1.82
     chart
    1.58
     tables
    1.55
    table
    1.40
     tabela
    1.33
     tabel
    1.30
     Table
    1.29
     Chart
    1.29
    chart
    1.25
     charts
    1.24
    Activations Density 0.516%

    No Known Activations