© 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-IT
    5. 31-GEMMASCOPE-2-RES-65K
    7. 55648
    Prev
    Next
    INDEX
    Explanations

    challenges and requirements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_65k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    is
    0.58
    ock
    0.55
    eran
    0.46
    ikos
    0.45
    ier
    0.44
    cado
    0.43
    ard
    0.42
    扶
    0.42
    ods
    0.41
    𝐞
    0.41
    POSITIVE LOGITS
     Finale
    0.57
     vinegar
    0.54
     finale
    0.49
     terminate
    0.48
    Finale
    0.47
    🍵
    0.46
     sauna
    0.45
    экс
    0.44
    ्यात
    0.43
    특별
    0.42
    Activations Density 0.004%

    No Known Activations