© 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. 7986
    Prev
    Next
    INDEX
    Explanations

    formal/legalistic language using `shall`

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

    shall, may, liable

    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
     murderous
    0.49
    Methylsulfanyl
    0.48
    变成了
    0.47
     پھی
    0.46
    ໋
    0.46
     henüz
    0.44
     אנחנו
    0.42
    馅
    0.42
     sultry
    0.42
    స్తోంది
    0.42
    POSITIVE LOGITS
     shall
    0.71
     pursuant
    0.56
    reasonable
    0.55
    provide
    0.54
     practicable
    0.51
    any
    0.51
     any
    0.50
     appropriate
    0.48
     may
    0.48
    shall
    0.48
    Activations Density 0.405%

    No Known Activations