© 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. 216
    Prev
    Next
    INDEX
    Explanations

    instances of quotation marks

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

    Follows "the"

    np_acts-logits-general · gemini-2.0-flash

    Fig. Exhibition names

    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
    <h2>
    -1.58
    <h3>
    -1.43
    </h2>
    -1.34
    <h4>
    -1.27
    <h1>
    -1.16
    <h5>
    -1.15
    ")){
    
    -1.11
    </h3>
    -1.10
    <h6>
    -1.04
    </h1>
    -1.04
    POSITIVE LOGITS
     Escolar
    0.83
     houſe
    0.81
     Houſe
    0.79
     Majefty
    0.74
     Artículos
    0.70
     purpoſe
    0.70
     brancas
    0.70
     figliu
    0.70
    vician
    0.70
     chré
    0.69
    Activations Density 0.040%

    No Known Activations