© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-65K
    7. 1719
    Prev
    Next
    INDEX
    Explanations

    core idea or argument

    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_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
     both
    0.17
    \
    0.17
     подхода
    0.17
            
    0.16
     various
    0.16
    $\
    0.16
             
    0.16
    0.16
               
    0.15
           
    0.15
    POSITIVE LOGITS
     annoyance
    0.20
     relegation
    0.18
     conundrum
    0.18
     indication
    0.18
     misalignment
    0.18
     inhomogeneity
    0.18
     requirement
    0.18
     inability
    0.18
     cynicism
    0.17
    પણે
    0.17
    Activations Density 2.013%

    No Known Activations