© 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-TRANSCODER-262K
    7. 49663
    Prev
    Next
    INDEX
    Explanations

    And followed by comma or conjunction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_31_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
     however
    0.39
     però
    0.35
     Однако
    0.35
     Ancak
    0.35
     그러나
    0.34
     W
    0.34
     However
    0.34
     HOWEVER
    0.34
    Однако
    0.33
     Cependant
    0.33
    POSITIVE LOGITS
    latin
    0.33
    '+
    0.29
     collusion
    0.29
    ន្ល
    0.28
     decouple
    0.26
    माल
    0.26
     bezeichnet
    0.26
     दशा
    0.26
     ensues
    0.26
    كان
    0.25
    Activations Density 0.002%

    No Known Activations