Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 2079
    Prev
    Next
    INDEX
    Explanations

    statistics related to baseball pitching performance

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

    strikeouts

    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
     ramah
    -0.54
    amaño
    -0.52
    Lähteet
    -0.51
     reklama
    -0.51
    vorschaubild
    -0.51
    ilhas
    -0.50
     miniaturka
    -0.50
     azeite
    -0.50
    válasz
    -0.50
    сылкі
    -0.49
    POSITIVE LOGITS
     strikeouts
    0.48
    يكب
    0.39
     pug
    0.36
    ]--;
    0.34
    RuleContext
    0.31
     prey
    0.31
    actéristiques
    0.31
     perfec
    0.30
    HasForeignKey
    0.30
    errHandler
    0.29
    Activations Density 0.005%

    No Known Activations