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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 30416
    Prev
    Next
    INDEX
    Explanations

    defense

    np_acts-logits-general · gemini-2.5-flash-lite

    descriptions of physical contact or positioning between players in sports, particularly focusing on movements past, around, or through defenders.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Patients
    -0.88
    伤
    -0.88
    LocalDate
    -0.87
    entech
    -0.85
    Patients
    -0.83
    perative
    -0.81
     ren
    -0.81
     crit
    -0.76
    ához
    -0.76
    复制代码
    -0.73
    POSITIVE LOGITS
     defenders
    3.23
     defenses
    2.53
     defender
    2.45
     defences
    2.44
     defense
    2.17
     defence
    2.03
     défense
    2.00
    defender
    1.97
     Defenders
    1.88
    defense
    1.80
    Activations Density 0.012%

    No Known Activations