© 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. 40-GEMMASCOPE-2-RES-16K
    7. 9593
    Prev
    Next
    INDEX
    Explanations

    accuracy in estimation and calibration

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

    prediction accuracy

    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_40_width_16k_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
     timeCounter
    0.48
     examining
    0.46
     लक्षण
    0.44
    检测
    0.43
    detectChanges
    0.42
    旷
    0.42
    探讨
    0.42
    遭遇
    0.41
     finalText
    0.41
     भेदभाव
    0.41
    POSITIVE LOGITS
     accuracy
    1.10
     approximations
    1.05
    accuracy
    1.04
     Accuracy
    1.04
     inaccuracy
    1.03
     error
    1.02
     approximation
    1.01
     calibration
    1.01
    Accuracy
    0.98
     accuracies
    0.98
    Activations Density 0.202%

    No Known Activations