© 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. 3899
    Prev
    Next
    INDEX
    Explanations

    retains, preserves, or keeps

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

    preserve or retain

    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
     भरो
    0.41
    क्षात्कार
    0.39
     tekrar
    0.38
     önce
    0.38
     ponownie
    0.37
     directement
    0.37
    ammlung
    0.37
    𝙛
    0.37
    正则
    0.37
     YOK
    0.37
    POSITIVE LOGITS
     retain
    1.16
    retain
    1.11
    保留
    1.06
     retains
    1.04
     retained
    1.02
     intact
    0.99
     unchanged
    0.98
     retention
    0.95
     сохранить
    0.89
     retaining
    0.87
    Activations Density 0.211%

    No Known Activations