© 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. 41658
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions followed by elaboration

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

    use of 'however' or 'but'

    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
    Об
    0.56
    о
    0.47
    оло
    0.46
    лга
    0.45
    є
    0.45
     ebenfalls
    0.43
    Де
    0.43
    І
    0.43
    До
    0.42
    પ્ર
    0.42
    POSITIVE LOGITS
     emojis
    0.51
     sát
    0.50
     আল্লাহ
    0.47
     skewers
    0.45
     Satoshi
    0.45
     spontaneity
    0.43
    を活用
    0.43
     playfully
    0.43
     parku
    0.42
     Beyoncé
    0.42
    Activations Density 0.000%

    No Known Activations