© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 119736
    Prev
    Next
    INDEX
    Explanations

    mentions of quarterback positions and related terms in the context of football

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

    football QB rating

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    </i>
    -0.50
    __*/
    -0.48
    princip
    -0.45
    typical
    -0.43
    <i>
    -0.43
     Astoria
    -0.42
    </em>
    -0.41
     Kars
    -0.40
     Yorkers
    -0.40
    orations
    -0.40
    POSITIVE LOGITS
    QB
    0.72
     gegen
    0.69
    qb
    0.68
    ReusableCell
    0.58
     qb
    0.58
    ChildScrollView
    0.57
     QB
    0.56
    Qb
    0.56
     FormBuilder
    0.55
     Houſe
    0.55
    Activations Density 0.002%

    No Known Activations