© 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-16K
    7. 9548
    Prev
    Next
    INDEX
    Explanations

    signals conclusion

    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_16k_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
     equating
    0.51
    calculate
    0.50
     Calculations
    0.50
    Calculating
    0.48
     Calculating
    0.47
     calculates
    0.47
     calcular
    0.47
    =((
    0.46
     calculate
    0.46
    Calcul
    0.46
    POSITIVE LOGITS
    结论
    0.38
     concl
    0.38
     vérifier
    0.36
     final
    0.36
    Verdict
    0.35
     förb
    0.35
     répondre
    0.35
    综
    0.35
    FINAL
    0.35
    CONCLUSION
    0.34
    Activations Density 0.281%

    No Known Activations