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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 8805
    Prev
    Next
    INDEX
    Explanations

    expressions of curiosity or amazement

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

    "wonder" followed by a question

    np_acts-logits-general · gemini-2.0-flash

    wonder if or whether

    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
    ########.
    -0.55
     noDo
    -0.53
    Shirts
    -0.52
    Sheikh
    -0.52
    cifix
    -0.52
    rshire
    -0.50
    Liam
    -0.50
    Aircraft
    -0.50
    Gloves
    -0.49
    Clik
    -0.49
    POSITIVE LOGITS
     WONDER
    0.70
     Wonderland
    0.66
     Wonder
    0.65
     Wonderful
    0.63
     Wunder
    0.60
     wonder
    0.57
     wonderland
    0.57
     wondered
    0.55
    wonder
    0.54
    quelize
    0.54
    Activations Density 0.125%

    No Known Activations