© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 1977
    Prev
    Next
    INDEX
    Explanations

    instances of noting or mentioning important information

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

    "note" followed by "that" or a word indicating possibility

    np_acts-logits-general · gemini-2.0-flash

    noting that

    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
    rshire
    -0.64
    انتهای
    -0.63
    "]);
    
    -0.62
     noDo
    -0.61
    arrête
    -0.59
    QObject
    -0.59
    glise
    -0.58
    '],$
    -0.58
    씩
    -0.57
    LVANIA
    -0.57
    POSITIVE LOGITS
     that
    0.94
    HtmlAttribute
    0.56
    日閲覧
    0.55
     estimés
    0.54
     חיצוניים
    0.54
     GetEnumerator
    0.51
    ثيق
    0.50
    VersionUID
    0.49
    Xna
    0.48
    ParallelGroup
    0.48
    Activations Density 0.536%

    No Known Activations