© 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. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 9069
    Prev
    Next
    INDEX
    Explanations

    words related to basketball teams, players, and associations

    oai_token-act-pair · gemini-2.0-flash

    Basketball Sacramento Kings

    np_max-act-logits · gemini-2.0-flash

    say "basketball" Method used: 2 — tokens after tend to be basketball-related, so say "basketball"

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     BASKET
    -1.63
     basketball
    -1.62
     court
    -1.50
     Basketball
    -1.48
     Basket
    -1.46
     basket
    -1.46
    Basketball
    -1.46
    Basket
    -1.41
    basket
    -1.40
    court
    -1.35
    POSITIVE LOGITS
     resourceCulture
    0.84
    eneuve
    0.61
    MigrationBuilder
    0.57
    AutoField
    0.57
    devamını
    0.56
     ::=
    0.56
     invokingState
    0.53
     Viitattu
    0.53
     StatefulWidget
    0.50
    Cyfeiriadau
    0.49
    Activations Density 1.530%

    No Known Activations