© 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. 24-GEMMASCOPE-2-TRANSCODER-262K
    7. 175536
    Prev
    Next
    INDEX
    Explanations

    dialogue punctuation

    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_24_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
    毕竟
    0.37
    Reddit
    0.37
     અમદાવાદ
    0.35
     reddit
    0.35
    道
    0.35
     ദൈ
    0.35
     해당
    0.34
    চরা
    0.33
     sonore
    0.33
     गिरफ्तार
    0.33
    POSITIVE LOGITS
     rong
    0.42
    Neck
    0.41
     Cio
    0.39
    eig
    0.38
     Neck
    0.37
    भ
    0.37
     quinze
    0.37
    Want
    0.36
     Eigen
    0.36
    Mis
    0.36
    Activations Density 0.000%

    No Known Activations