© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 189751
    Prev
    Next
    INDEX
    Explanations

    police constable cursing rejecting

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

    trachea rings, CRF, rhythm, chase, quintic, finding large, stage, teams

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    המ
    0.55
    垦
    0.45
     eins
    0.44
    Brexit
    0.43
    Service
    0.43
    Environmental
    0.42
    뜯
    0.42
     agric
    0.42
    負
    0.41
     redox
    0.41
    POSITIVE LOGITS
     પોલીસે
    0.52
     cursing
    0.50
     प्रहरी
    0.49
    ಡಿದ
    0.47
    нь
    0.47
    qiya
    0.46
     rejecting
    0.46
     constable
    0.45
     scrubbing
    0.45
    ৃষ্ট
    0.45
    Activations Density 0.001%

    No Known Activations