© 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. 57-GEMMASCOPE-2-TRANSCODER-262K
    7. 39633
    Prev
    Next
    INDEX
    Explanations

    say "similarity or equality"

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_57_width_262k_l0_small_affine
    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
     closely
    0.51
    朗
    0.40
     বিস্তৃত
    0.40
     curios
    0.39
     ouverture
    0.39
     curious
    0.38
    经销商
    0.38
     perfectly
    0.38
    +)$
    0.38
     kudos
    0.38
    POSITIVE LOGITS
     equal
    0.90
     igual
    0.88
    Equal
    0.84
     equaling
    0.82
     Equal
    0.81
     рав
    0.80
     equals
    0.79
    equal
    0.79
     correspond
    0.70
     égale
    0.70
    Activations Density 0.018%

    No Known Activations