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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 91378
    Prev
    Next
    INDEX
    Explanations

    words and phrases associated with farewells and parting

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

    saying goodbyefarewellsgoodbyes

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    eseorang
    -0.55
    IntoConstraints
    -0.47
    Pranala
    -0.41
    Бахар
    -0.36
     useRouter
    -0.36
    staaten
    -0.36
    Geograf
    -0.34
    //*/
    -0.33
    guém
    -0.33
    Council
    -0.32
    POSITIVE LOGITS
     goodbye
    1.23
     farewell
    1.20
    goodbye
    1.15
     Goodbye
    1.09
     Farewell
    1.05
    Goodbye
    1.05
    Farewell
    1.01
    byes
    0.99
    bye
    0.90
    Bye
    0.86
    Activations Density 0.005%

    No Known Activations