© 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. 27026
    Prev
    Next
    INDEX
    Explanations

    experience and alignment

    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
     exacerb
    0.45
    வதால்
    0.41
     komplik
    0.41
    เปลี่ยนแปลง
    0.40
     modifies
    0.39
     ойнойт
    0.39
    ၏
    0.39
    ('\
    0.39
     implication
    0.38
     lanceol
    0.37
    POSITIVE LOGITS
     earning
    0.64
     earn
    0.61
     earns
    0.58
     impressed
    0.54
     impress
    0.53
     earned
    0.51
     impresses
    0.51
     Earn
    0.50
     culminated
    0.48
     মুগ্ধ
    0.48
    Activations Density 0.003%

    No Known Activations