© 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. 40-GEMMASCOPE-2-RES-262K
    7. 131526
    Prev
    Next
    INDEX
    Explanations

    being aware and understanding

    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_40_width_262k_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
     bipartisan
    0.48
      
    0.42
    Fix
    0.40
    Political
    0.40
    House
    0.39
     رجال
    0.39
    Cu
    0.39
    冗
    0.38
    BD
    0.37
    readme
    0.37
    POSITIVE LOGITS
     pyramidal
    0.46
    暗
    0.44
     graham
    0.44
     klik
    0.43
     dark
    0.42
     lampu
    0.41
     estern
    0.41
     lumière
    0.41
     inqu
    0.41
     migli
    0.41
    Activations Density 0.003%

    No Known Activations