© 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. 15012
    Prev
    Next
    INDEX
    Explanations

    actions with negative intent

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

    intentional or malicious intent

    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
    Streams
    0.43
     ஏரா
    0.42
    好評
    0.41
    茫
    0.39
    ಗಳಲ್ಲಿ
    0.39
    ಗಳಿಂದ
    0.39
    olefin
    0.39
     streams
    0.38
     ::=
    0.38
     Certificates
    0.38
    POSITIVE LOGITS
     intentionally
    1.20
     purposely
    1.17
    故意
    1.13
     purposefully
    1.09
     deliberately
    1.07
     intentional
    1.04
     साजिश
    0.98
     conspiracy
    0.92
     ষড়যন্ত্র
    0.89
    intentional
    0.88
    Activations Density 0.117%

    No Known Activations