© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 4-GEMMASCOPE-2-TRANSCODER-262K
    7. 194987
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_4_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
    ي
    0.63
    و
    0.55
    зан
    0.53
    Potion
    0.52
    jším
    0.51
    وجه
    0.50
    ра
    0.50
    舉辦
    0.50
    اي
    0.49
    鹘
    0.49
    POSITIVE LOGITS
     synapses
    0.56
     receptors
    0.55
     reddit
    0.54
     oblig
    0.54
     slats
    0.54
     Restaurants
    0.53
     reviews
    0.52
     hostels
    0.52
     m
    0.52
     credits
    0.52
    Activations Density 0.000%

    No Known Activations