© 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. 31-GEMMASCOPE-2-RES-262K
    7. 48033
    Prev
    Next
    INDEX
    Explanations

    always is the spice of life

    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_31_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
     دیگر
    0.46
     Dude
    0.42
    dude
    0.41
    idon
    0.41
     doping
    0.41
     exhausted
    0.39
     abound
    0.39
     wg
    0.39
    immon
    0.39
     pula
    0.38
    POSITIVE LOGITS
     jednotliv
    0.51
     実
    0.44
    䡇
    0.43
     vedno
    0.43
     vždy
    0.42
    ясплат
    0.42
    Always
    0.41
     különböz
    0.41
    केवल
    0.41
     পাশ
    0.40
    Activations Density 0.008%

    No Known Activations