© 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. 50-GEMMASCOPE-2-TRANSCODER-262K
    7. 200461
    Prev
    Next
    INDEX
    Explanations

    chip or nano

    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_50_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
     রেলপথ
    0.49
    Ղ
    0.45
    REN
    0.41
    ధాన
    0.39
     Gunnar
    0.39
     lidar
    0.38
     mountain
    0.37
     Arn
    0.35
     règle
    0.35
    規
    0.35
    POSITIVE LOGITS
     Nand
    0.45
    
    0.45
    chip
    0.44
    Nano
    0.44
     beacons
    0.43
     Characteristics
    0.43
    nano
    0.42
    Pis
    0.42
    ピ
    0.41
     Chip
    0.41
    Activations Density 0.005%

    No Known Activations