{ "asset_sha256": { "data/ben_Beng.devtest": "6699aa77b4c93d520971868cd5ff06a1e3b5ddfde852da13337f5194bc23086f", "data/eng_Latn.devtest": "612e9fbe87997617c0fa8fa8929654a4f49b728d96738112c2b86ef6a1d78d88", "data/hin_Deva.devtest": "5f5fd39acadca29fb044a0398e81869e48f37de979df086f2fad4a7d1fd2d015", "data/mar_Deva.devtest": "63d0d1bfdcfa7f8a1538c87cab65b0e5e010515529ed286b1f8b2a51cc62b99f", "data/tam_Taml.devtest": "a18b26bf278e458f085c70173d6789d6108cb2e3346c81a1bd0f8aceb4ea30e4", "data/tel_Telu.devtest": "2dc641b4fb69347efe4b0c0f2063372c643f20c664b41a1f4421b69ba2a06420", "tokenizers/GPT-2.json": "8414cab924d8b9b33013f0d221c5862f365ee9be39c5c2bfae8a5a9e970478a6", "tokenizers/Qwen2.5.json": "c0382117ea329cdf097041132f6d735924b697924d6f6fc3945713e96ce87539", "tokenizers/UMT5.json": "af904105ce1071b1202bba0059a841f4a7b85b48b6ec179c4948e3483476e0dd", "tokenizers/XLM-R.json": "a898ea75433890f6610f4e470b8ebeb0c21dce5c8dd61f892eb09eb5919d2e2c", "tokenizers/mBERT.json": "f4a4d5bf7301717e261fafbe26e1eb967f6ba4cb3ae0ab7a29f4642ec229f386" }, "bootstrap_resamples": 2000, "bootstrap_unit": "aligned sentence row", "context_tokens": 4096, "counts": { "GPT-2": { "Bengali": [ 344, 599, 522, 157, 363, 196, 293, 386, 520, 248, 274, 390, 218, 124, 236, 134, 257, 430, 451, 259, 202, 264, 400, 217, 351, 344, 390, 132, 199, 250, 308, 248, 135, 306, 123, 191, 131, 417, 166, 186, 120, 369, 312, 168, 304, 196, 298, 262, 186, 199, 180, 281, 275, 288, 331, 382, 184, 157, 627, 273, 348, 189, 363, 238, 262, 203, 168, 360, 456, 288, 118, 131, 196, 234, 244, 169, 191, 291, 206, 93, 333, 311, 278, 188, 142, 354, 201, 492, 189, 177, 149, 147, 305, 364, 365, 72, 179, 224, 300, 330, 406, 233, 221, 285, 306, 287, 681, 304, 169, 183, 226, 193, 119, 266, 202, 182, 214, 221, 311, 269, 341, 169, 212, 204, 232, 344, 242, 78, 133, 109, 144, 270, 207, 366, 515, 118, 348, 270, 191, 382, 200, 281, 134, 221, 261, 139, 325, 337, 258, 148, 174, 166, 241, 267, 187, 331, 196, 519, 216, 377, 206, 270, 185, 202, 354, 243, 224, 195, 327, 228, 260, 282, 430, 434, 392, 334, 351, 357, 347, 215, 164, 349, 147, 253, 349, 201, 303, 301, 203, 212, 335, 266, 152, 192, 165, 92, 163, 220, 228, 217, 255, 375, 390, 141, 218, 245, 337, 343, 171, 220, 383, 229, 480, 369, 168, 178, 183, 276, 330, 280, 269, 283, 217, 331, 185, 298, 160, 239, 206, 254, 260, 360, 169, 213, 269, 237, 373, 360, 205, 163, 173, 211, 236, 285, 179, 262, 260, 354, 135, 309, 164, 273, 274, 92, 408, 286, 310, 375, 191, 261, 254, 303, 138, 247, 202, 284, 257, 148, 202, 190, 277, 250, 338, 357, 170, 293, 260, 174, 252, 346, 218, 335, 230, 256, 276, 429, 325, 213, 250, 145, 355, 243, 140, 130, 233, 131, 245, 245, 287, 234 ], "English": [ 31, 60, 45, 14, 40, 18, 23, 33, 49, 29, 27, 39, 24, 14, 28, 14, 27, 37, 39, 28, 19, 33, 34, 27, 31, 37, 37, 13, 23, 22, 33, 23, 11, 34, 11, 20, 13, 37, 18, 25, 15, 31, 31, 14, 39, 16, 30, 25, 21, 24, 19, 29, 27, 21, 27, 39, 19, 16, 63, 30, 32, 13, 40, 26, 33, 16, 20, 41, 53, 31, 21, 13, 16, 23, 25, 25, 32, 30, 25, 15, 35, 29, 25, 27, 16, 40, 17, 72, 30, 19, 17, 14, 24, 27, 35, 12, 18, 22, 38, 40, 35, 23, 21, 24, 29, 29, 68, 41, 18, 22, 23, 26, 17, 29, 24, 19, 36, 30, 41, 30, 32, 18, 27, 17, 26, 43, 19, 12, 16, 14, 17, 37, 20, 33, 38, 14, 29, 20, 29, 50, 29, 32, 21, 30, 21, 15, 37, 44, 23, 20, 28, 22, 27, 26, 18, 30, 22, 35, 19, 34, 25, 22, 22, 20, 40, 23, 20, 19, 36, 22, 22, 32, 40, 48, 40, 28, 44, 35, 40, 21, 19, 36, 15, 24, 33, 22, 27, 25, 18, 21, 30, 32, 14, 21, 12, 9, 19, 21, 24, 28, 15, 35, 38, 18, 20, 22, 29, 32, 13, 19, 36, 24, 45, 37, 18, 23, 21, 30, 32, 31, 33, 39, 20, 35, 24, 33, 21, 22, 24, 23, 30, 34, 29, 21, 29, 24, 30, 31, 22, 18, 21, 25, 37, 32, 18, 29, 25, 30, 20, 37, 21, 31, 34, 12, 36, 22, 37, 26, 22, 28, 25, 25, 17, 28, 18, 24, 29, 20, 22, 29, 24, 24, 37, 37, 25, 35, 27, 18, 27, 36, 27, 30, 21, 27, 38, 42, 32, 22, 24, 22, 37, 15, 25, 15, 28, 14, 26, 29, 30, 27 ], "Hindi": [ 308, 466, 370, 135, 264, 145, 233, 304, 367, 188, 211, 276, 194, 102, 160, 86, 235, 299, 329, 195, 133, 214, 300, 185, 250, 277, 261, 98, 153, 162, 209, 208, 110, 181, 112, 118, 99, 321, 128, 127, 60, 287, 236, 103, 256, 133, 210, 264, 115, 142, 116, 195, 210, 233, 191, 301, 138, 104, 469, 168, 286, 131, 199, 154, 208, 139, 136, 290, 330, 231, 162, 99, 156, 184, 177, 120, 121, 262, 165, 111, 247, 231, 247, 199, 87, 304, 138, 388, 124, 114, 102, 125, 205, 278, 209, 66, 176, 211, 272, 248, 288, 220, 180, 198, 253, 196, 537, 313, 127, 134, 146, 155, 130, 205, 140, 149, 188, 169, 287, 204, 247, 133, 165, 118, 193, 343, 180, 63, 82, 74, 124, 197, 155, 241, 318, 127, 277, 227, 183, 284, 159, 204, 131, 184, 160, 128, 279, 284, 160, 113, 181, 150, 187, 167, 142, 252, 123, 315, 193, 329, 208, 185, 183, 175, 269, 214, 136, 158, 278, 180, 166, 210, 295, 414, 276, 200, 259, 302, 247, 164, 144, 245, 130, 244, 259, 132, 235, 220, 178, 240, 255, 187, 128, 143, 110, 70, 157, 199, 186, 122, 162, 335, 319, 103, 184, 225, 226, 283, 137, 174, 289, 167, 426, 322, 162, 124, 173, 279, 258, 221, 191, 227, 123, 223, 133, 220, 118, 173, 181, 165, 213, 227, 119, 141, 213, 188, 244, 270, 165, 137, 119, 201, 193, 232, 183, 247, 167, 282, 78, 317, 146, 236, 237, 104, 312, 189, 245, 289, 171, 194, 201, 193, 123, 219, 170, 254, 251, 113, 166, 177, 162, 178, 273, 254, 165, 233, 146, 139, 206, 250, 229, 263, 175, 201, 241, 341, 228, 138, 118, 160, 298, 220, 123, 90, 174, 112, 217, 199, 237, 184 ], "Marathi": [ 207, 565, 389, 133, 310, 174, 192, 335, 384, 199, 211, 318, 191, 84, 190, 90, 247, 351, 365, 163, 163, 188, 323, 229, 283, 350, 292, 118, 187, 165, 218, 255, 118, 211, 110, 147, 110, 383, 119, 177, 80, 261, 249, 103, 275, 160, 239, 210, 134, 167, 145, 237, 236, 243, 259, 307, 170, 110, 492, 219, 271, 128, 258, 140, 270, 148, 158, 303, 305, 208, 128, 121, 172, 177, 167, 115, 134, 262, 163, 100, 260, 228, 234, 194, 89, 294, 157, 432, 131, 125, 121, 117, 216, 213, 264, 63, 163, 217, 250, 298, 307, 209, 132, 242, 247, 236, 547, 323, 116, 132, 190, 158, 133, 230, 156, 156, 186, 215, 268, 260, 311, 151, 188, 129, 208, 356, 153, 80, 100, 94, 146, 251, 163, 261, 320, 110, 299, 217, 171, 305, 158, 225, 161, 202, 191, 135, 274, 261, 188, 110, 154, 103, 195, 195, 124, 263, 143, 317, 158, 343, 190, 197, 165, 157, 290, 198, 165, 149, 206, 145, 216, 200, 287, 365, 247, 229, 258, 301, 239, 149, 143, 289, 120, 251, 256, 171, 236, 217, 182, 178, 231, 217, 109, 150, 127, 96, 184, 162, 185, 177, 143, 318, 363, 136, 179, 191, 209, 239, 174, 197, 298, 232, 420, 352, 169, 144, 233, 218, 238, 221, 231, 203, 170, 269, 165, 251, 125, 151, 154, 170, 207, 239, 140, 153, 242, 178, 240, 264, 175, 132, 183, 207, 237, 247, 156, 242, 227, 248, 109, 265, 136, 228, 238, 90, 298, 189, 261, 221, 176, 217, 211, 235, 131, 192, 163, 202, 234, 160, 168, 186, 211, 187, 349, 255, 147, 259, 163, 118, 211, 218, 180, 248, 199, 244, 213, 397, 233, 166, 155, 164, 329, 165, 144, 108, 185, 127, 219, 203, 237, 186 ], "Tamil": [ 536, 875, 848, 273, 611, 341, 464, 637, 754, 514, 537, 746, 431, 201, 407, 154, 438, 734, 600, 420, 306, 402, 627, 375, 491, 602, 531, 183, 351, 289, 357, 415, 211, 497, 234, 373, 187, 793, 326, 305, 204, 638, 444, 165, 590, 262, 530, 365, 251, 376, 291, 448, 402, 485, 663, 630, 325, 242, 915, 364, 475, 283, 643, 358, 347, 254, 315, 541, 747, 462, 249, 224, 398, 421, 316, 294, 212, 515, 340, 205, 609, 596, 449, 466, 130, 546, 278, 766, 337, 241, 258, 217, 416, 442, 631, 136, 284, 311, 412, 661, 566, 308, 335, 332, 570, 415, 1122, 638, 320, 274, 434, 372, 375, 488, 353, 330, 386, 307, 618, 369, 602, 318, 465, 291, 308, 713, 379, 194, 248, 194, 236, 504, 365, 483, 598, 270, 587, 479, 311, 554, 328, 398, 173, 312, 353, 248, 589, 582, 357, 223, 333, 333, 359, 363, 248, 481, 311, 638, 328, 525, 426, 442, 304, 366, 503, 388, 379, 309, 511, 425, 316, 428, 594, 970, 599, 490, 576, 529, 564, 346, 297, 459, 262, 497, 499, 309, 446, 488, 324, 409, 575, 419, 279, 277, 284, 181, 280, 377, 337, 380, 341, 572, 526, 289, 277, 413, 480, 467, 237, 380, 550, 526, 686, 652, 375, 252, 370, 480, 474, 388, 451, 403, 312, 625, 354, 521, 270, 397, 301, 438, 634, 571, 289, 360, 394, 316, 485, 555, 382, 291, 372, 377, 458, 516, 284, 597, 391, 618, 208, 624, 294, 514, 576, 184, 570, 421, 551, 517, 304, 426, 582, 509, 231, 447, 314, 408, 389, 235, 268, 429, 418, 424, 535, 541, 292, 523, 321, 244, 356, 662, 439, 526, 424, 425, 425, 708, 526, 354, 366, 303, 591, 335, 307, 213, 293, 237, 441, 346, 482, 350 ], "Telugu": [ 507, 698, 534, 214, 537, 248, 393, 545, 640, 319, 417, 469, 317, 173, 237, 179, 432, 554, 601, 267, 301, 338, 518, 232, 357, 486, 379, 104, 287, 263, 387, 343, 182, 402, 225, 301, 179, 451, 221, 243, 156, 552, 499, 156, 397, 257, 468, 325, 220, 338, 314, 355, 298, 321, 385, 519, 261, 202, 962, 314, 499, 189, 451, 300, 448, 269, 301, 489, 618, 322, 202, 195, 224, 317, 280, 247, 237, 403, 309, 147, 437, 427, 332, 321, 119, 465, 238, 753, 255, 214, 189, 190, 381, 437, 490, 137, 273, 277, 492, 479, 530, 360, 253, 330, 505, 315, 968, 600, 255, 215, 329, 301, 299, 314, 240, 335, 294, 244, 471, 344, 520, 205, 326, 276, 236, 493, 323, 136, 177, 139, 233, 381, 265, 453, 345, 216, 340, 352, 283, 527, 133, 333, 141, 318, 286, 211, 507, 574, 292, 107, 336, 424, 306, 279, 226, 455, 272, 469, 397, 546, 284, 344, 199, 242, 450, 302, 287, 260, 446, 406, 300, 346, 507, 592, 432, 408, 415, 490, 459, 298, 225, 441, 253, 390, 438, 233, 386, 338, 312, 319, 450, 340, 284, 260, 216, 116, 210, 225, 331, 277, 363, 539, 569, 205, 268, 425, 411, 492, 248, 362, 494, 428, 739, 567, 272, 259, 300, 417, 402, 349, 442, 349, 190, 440, 270, 399, 246, 334, 272, 317, 387, 466, 234, 273, 429, 297, 420, 469, 343, 259, 272, 334, 333, 383, 259, 470, 404, 328, 138, 523, 187, 429, 432, 168, 545, 316, 400, 452, 226, 417, 371, 378, 215, 352, 271, 446, 380, 238, 277, 315, 308, 354, 379, 444, 246, 350, 313, 216, 382, 339, 256, 476, 310, 321, 425, 569, 467, 314, 179, 211, 363, 320, 244, 176, 312, 164, 320, 348, 393, 330 ] }, "Qwen2.5": { "Bengali": [ 184, 321, 272, 83, 198, 101, 162, 205, 287, 140, 152, 199, 114, 64, 128, 71, 134, 227, 237, 135, 109, 142, 217, 116, 185, 174, 208, 68, 105, 128, 174, 130, 71, 176, 65, 92, 67, 228, 91, 104, 63, 190, 160, 86, 165, 103, 156, 138, 111, 104, 98, 148, 149, 147, 175, 201, 97, 88, 330, 156, 190, 100, 196, 129, 138, 110, 92, 186, 239, 163, 68, 69, 109, 131, 133, 90, 102, 152, 112, 49, 183, 174, 153, 96, 74, 192, 109, 281, 101, 99, 75, 73, 160, 187, 196, 38, 97, 121, 162, 186, 215, 124, 122, 159, 167, 156, 353, 168, 92, 106, 118, 103, 61, 145, 110, 99, 114, 119, 168, 147, 187, 93, 113, 108, 126, 193, 124, 39, 70, 59, 76, 148, 113, 202, 283, 62, 184, 143, 103, 217, 111, 148, 76, 129, 135, 79, 174, 174, 134, 79, 101, 91, 125, 142, 99, 178, 106, 284, 116, 204, 108, 142, 100, 101, 191, 132, 119, 110, 180, 122, 135, 147, 230, 232, 207, 182, 195, 189, 186, 114, 82, 192, 76, 131, 188, 105, 157, 156, 111, 114, 173, 141, 82, 103, 90, 49, 84, 118, 122, 116, 142, 196, 206, 78, 116, 133, 178, 176, 86, 115, 203, 121, 249, 201, 90, 93, 96, 149, 173, 149, 144, 157, 117, 178, 98, 160, 86, 127, 103, 135, 146, 193, 101, 116, 138, 126, 197, 192, 113, 90, 94, 112, 137, 151, 103, 143, 142, 188, 74, 168, 88, 145, 151, 49, 209, 149, 162, 195, 101, 142, 140, 163, 74, 132, 110, 156, 129, 86, 111, 106, 147, 132, 187, 200, 94, 163, 140, 89, 137, 180, 116, 183, 119, 135, 155, 235, 176, 113, 137, 78, 195, 132, 72, 68, 132, 70, 124, 136, 150, 129 ], "English": [ 32, 63, 46, 14, 42, 20, 24, 32, 49, 29, 28, 39, 24, 14, 28, 17, 27, 38, 42, 28, 19, 33, 34, 27, 31, 38, 39, 12, 23, 22, 34, 23, 11, 43, 11, 19, 13, 37, 20, 27, 15, 31, 30, 19, 40, 16, 30, 23, 25, 23, 18, 28, 31, 21, 27, 39, 19, 15, 63, 31, 36, 13, 40, 26, 33, 16, 20, 39, 51, 36, 25, 13, 16, 24, 26, 25, 29, 28, 22, 15, 36, 31, 24, 26, 16, 44, 17, 79, 29, 19, 17, 14, 24, 27, 35, 12, 18, 26, 38, 39, 39, 24, 21, 26, 29, 29, 68, 40, 18, 22, 23, 27, 17, 29, 24, 19, 35, 30, 41, 30, 32, 18, 27, 17, 25, 44, 19, 12, 16, 14, 17, 41, 26, 31, 38, 14, 29, 20, 28, 62, 30, 34, 27, 29, 21, 16, 39, 42, 23, 20, 28, 22, 27, 30, 18, 30, 26, 35, 19, 34, 25, 22, 22, 20, 39, 22, 19, 20, 35, 22, 22, 32, 41, 49, 40, 27, 47, 35, 39, 21, 19, 41, 14, 25, 32, 22, 28, 23, 18, 20, 31, 32, 14, 21, 12, 9, 17, 21, 24, 28, 15, 35, 36, 18, 20, 23, 29, 32, 13, 19, 35, 23, 45, 41, 18, 23, 21, 30, 33, 33, 33, 39, 21, 35, 23, 34, 18, 21, 24, 23, 33, 34, 32, 24, 29, 22, 33, 33, 22, 18, 20, 25, 36, 31, 18, 28, 26, 30, 20, 36, 21, 31, 39, 12, 36, 22, 37, 26, 22, 27, 25, 25, 17, 30, 18, 24, 28, 20, 27, 28, 24, 24, 37, 37, 24, 35, 26, 18, 27, 36, 27, 30, 21, 27, 36, 43, 32, 22, 22, 22, 37, 16, 24, 15, 27, 15, 25, 29, 30, 28 ], "Hindi": [ 182, 271, 223, 75, 162, 87, 143, 180, 230, 116, 125, 155, 123, 63, 102, 52, 132, 177, 193, 114, 79, 131, 174, 110, 141, 165, 155, 59, 100, 100, 146, 123, 66, 124, 68, 70, 56, 187, 80, 79, 38, 168, 131, 70, 151, 81, 134, 149, 81, 83, 68, 114, 137, 145, 123, 179, 88, 64, 285, 107, 182, 83, 120, 89, 123, 88, 81, 170, 194, 147, 99, 57, 93, 108, 107, 69, 74, 151, 98, 65, 157, 143, 137, 123, 54, 189, 83, 244, 77, 73, 62, 70, 122, 167, 122, 38, 98, 125, 170, 151, 175, 135, 107, 121, 153, 116, 306, 196, 80, 84, 86, 85, 77, 130, 90, 93, 116, 110, 174, 126, 147, 84, 103, 77, 122, 212, 102, 42, 53, 44, 77, 123, 96, 139, 192, 77, 155, 123, 110, 184, 96, 133, 89, 112, 103, 76, 176, 172, 97, 72, 116, 94, 105, 107, 81, 155, 77, 201, 117, 197, 132, 107, 116, 103, 162, 133, 82, 96, 162, 115, 102, 124, 177, 259, 163, 123, 153, 180, 148, 101, 89, 151, 75, 148, 153, 84, 127, 132, 102, 141, 153, 115, 79, 91, 66, 40, 94, 122, 118, 75, 93, 206, 190, 65, 111, 130, 132, 170, 80, 99, 172, 99, 259, 195, 103, 71, 107, 173, 157, 145, 114, 148, 72, 143, 82, 137, 73, 101, 106, 101, 133, 147, 80, 90, 125, 112, 152, 172, 99, 85, 71, 130, 117, 136, 103, 145, 106, 172, 46, 187, 88, 151, 146, 60, 182, 112, 151, 168, 94, 113, 124, 120, 72, 140, 105, 153, 154, 73, 105, 111, 103, 109, 167, 162, 108, 151, 94, 76, 126, 152, 143, 162, 107, 117, 152, 209, 139, 88, 71, 100, 176, 126, 77, 58, 108, 66, 122, 124, 140, 103 ], "Marathi": [ 125, 327, 232, 77, 182, 101, 116, 204, 234, 124, 128, 191, 109, 50, 113, 60, 150, 209, 217, 94, 98, 111, 189, 141, 166, 203, 172, 69, 115, 102, 133, 144, 74, 138, 62, 86, 64, 224, 76, 110, 49, 154, 145, 68, 162, 101, 147, 129, 85, 96, 86, 137, 144, 138, 159, 186, 103, 67, 298, 135, 167, 76, 155, 86, 156, 89, 93, 179, 174, 133, 83, 68, 108, 107, 99, 65, 84, 153, 102, 62, 157, 135, 131, 114, 55, 183, 91, 261, 81, 79, 71, 68, 122, 125, 153, 38, 96, 124, 149, 176, 182, 127, 82, 144, 150, 133, 310, 195, 72, 81, 112, 90, 82, 138, 96, 94, 114, 134, 163, 158, 182, 95, 114, 76, 124, 213, 93, 49, 60, 53, 92, 159, 100, 156, 193, 65, 170, 122, 106, 195, 100, 136, 102, 123, 119, 76, 168, 159, 112, 67, 101, 65, 113, 116, 76, 154, 91, 187, 92, 194, 114, 109, 96, 97, 171, 123, 101, 88, 120, 88, 130, 120, 176, 222, 147, 144, 157, 173, 147, 91, 85, 177, 68, 147, 152, 105, 131, 124, 104, 113, 135, 127, 66, 91, 78, 56, 107, 97, 109, 112, 82, 199, 213, 83, 103, 108, 117, 149, 103, 117, 174, 138, 248, 208, 101, 82, 136, 136, 151, 140, 134, 127, 95, 162, 102, 149, 76, 89, 90, 109, 132, 148, 95, 95, 139, 106, 147, 154, 106, 85, 108, 126, 143, 143, 95, 144, 144, 144, 64, 161, 83, 139, 142, 51, 171, 116, 156, 132, 104, 128, 128, 140, 73, 116, 96, 121, 135, 97, 105, 116, 126, 108, 212, 157, 91, 155, 103, 64, 128, 135, 106, 151, 117, 135, 127, 232, 138, 100, 89, 99, 192, 96, 94, 66, 117, 78, 123, 124, 144, 107 ], "Tamil": [ 214, 354, 330, 107, 249, 134, 180, 253, 293, 202, 210, 290, 165, 79, 158, 68, 176, 306, 240, 167, 126, 159, 245, 161, 200, 246, 214, 77, 141, 115, 146, 166, 86, 201, 92, 148, 72, 318, 132, 124, 85, 256, 175, 74, 233, 100, 208, 147, 100, 150, 117, 182, 174, 195, 264, 249, 132, 94, 374, 154, 196, 110, 247, 149, 148, 102, 127, 217, 302, 192, 104, 85, 164, 174, 137, 120, 89, 200, 139, 81, 244, 237, 181, 190, 53, 220, 109, 331, 142, 100, 99, 82, 164, 177, 247, 55, 115, 127, 174, 269, 233, 126, 133, 141, 229, 161, 441, 259, 128, 110, 175, 144, 150, 187, 139, 128, 158, 124, 243, 145, 243, 130, 179, 118, 127, 284, 146, 73, 100, 76, 92, 202, 153, 195, 243, 108, 230, 182, 126, 243, 130, 165, 76, 127, 138, 97, 235, 230, 143, 86, 138, 133, 144, 149, 101, 191, 130, 257, 132, 208, 167, 179, 122, 143, 203, 155, 151, 124, 211, 173, 127, 169, 242, 375, 233, 186, 230, 211, 224, 135, 117, 189, 101, 191, 202, 123, 176, 195, 131, 163, 231, 163, 112, 107, 109, 70, 112, 145, 133, 150, 133, 230, 219, 110, 112, 158, 199, 183, 94, 151, 217, 203, 275, 257, 150, 97, 146, 192, 196, 163, 184, 178, 123, 255, 142, 209, 109, 164, 122, 175, 250, 236, 126, 145, 159, 131, 198, 219, 153, 115, 149, 147, 185, 207, 113, 240, 158, 249, 86, 242, 118, 213, 236, 75, 233, 173, 225, 206, 120, 173, 231, 197, 86, 184, 133, 157, 157, 94, 108, 168, 167, 165, 218, 220, 121, 209, 129, 96, 146, 260, 174, 213, 164, 170, 177, 288, 212, 134, 150, 124, 239, 139, 124, 85, 117, 96, 171, 139, 193, 141 ], "Telugu": [ 280, 381, 293, 117, 306, 139, 216, 302, 347, 173, 225, 253, 176, 90, 125, 103, 233, 310, 337, 143, 160, 187, 285, 131, 198, 266, 203, 60, 160, 143, 209, 185, 98, 231, 119, 177, 101, 245, 122, 131, 86, 304, 268, 91, 219, 138, 249, 177, 128, 189, 166, 200, 178, 176, 218, 282, 142, 113, 529, 184, 277, 101, 248, 172, 249, 144, 163, 273, 341, 181, 114, 108, 126, 173, 157, 135, 129, 222, 171, 82, 236, 240, 189, 179, 68, 260, 131, 445, 139, 118, 106, 103, 210, 241, 282, 72, 152, 157, 274, 268, 287, 196, 137, 176, 278, 171, 537, 327, 144, 114, 180, 159, 162, 172, 135, 183, 162, 133, 260, 186, 279, 112, 173, 150, 131, 266, 173, 72, 100, 77, 126, 209, 151, 244, 190, 120, 185, 188, 160, 307, 84, 185, 82, 178, 160, 115, 274, 313, 162, 58, 187, 233, 168, 157, 124, 238, 155, 258, 214, 296, 153, 191, 112, 134, 251, 163, 156, 142, 242, 223, 159, 189, 278, 314, 240, 221, 230, 267, 255, 163, 123, 253, 141, 211, 244, 131, 214, 178, 174, 178, 248, 194, 154, 143, 122, 63, 113, 125, 177, 147, 197, 298, 304, 110, 151, 229, 220, 268, 132, 195, 279, 231, 405, 317, 153, 144, 169, 230, 216, 199, 238, 203, 104, 240, 149, 222, 134, 184, 153, 181, 210, 250, 139, 156, 232, 162, 234, 257, 188, 147, 151, 189, 184, 215, 144, 255, 218, 188, 77, 282, 103, 240, 239, 97, 302, 175, 220, 246, 130, 225, 207, 206, 117, 196, 145, 242, 205, 130, 158, 171, 163, 194, 204, 251, 138, 190, 176, 116, 211, 187, 142, 265, 175, 172, 230, 310, 260, 178, 104, 118, 202, 171, 135, 100, 171, 92, 170, 190, 215, 176 ] }, "UMT5": { "Bengali": [ 62, 114, 91, 29, 79, 40, 51, 66, 98, 51, 49, 61, 40, 20, 48, 28, 45, 85, 77, 43, 38, 51, 70, 45, 53, 61, 62, 24, 47, 42, 67, 45, 24, 74, 25, 33, 26, 71, 37, 41, 17, 64, 55, 26, 53, 31, 55, 46, 44, 44, 33, 47, 62, 46, 56, 64, 34, 31, 127, 54, 71, 30, 73, 49, 61, 36, 33, 66, 91, 62, 29, 20, 34, 45, 47, 33, 39, 58, 40, 14, 68, 52, 54, 32, 28, 71, 39, 118, 39, 34, 25, 22, 45, 63, 76, 13, 28, 34, 56, 66, 64, 49, 41, 52, 59, 50, 111, 76, 36, 37, 40, 36, 22, 50, 36, 34, 47, 47, 68, 48, 56, 33, 38, 36, 46, 71, 45, 17, 21, 23, 25, 65, 52, 71, 87, 22, 56, 41, 39, 91, 41, 58, 37, 45, 48, 31, 60, 55, 44, 31, 43, 36, 49, 52, 37, 53, 37, 101, 42, 71, 34, 47, 32, 33, 71, 43, 40, 33, 69, 40, 42, 57, 74, 82, 69, 60, 70, 61, 64, 42, 37, 67, 24, 49, 64, 32, 56, 51, 39, 44, 63, 55, 26, 37, 26, 19, 33, 40, 44, 43, 45, 54, 67, 25, 39, 45, 55, 65, 29, 35, 65, 41, 79, 77, 32, 36, 34, 51, 61, 56, 52, 64, 41, 60, 32, 58, 28, 43, 36, 53, 56, 66, 41, 42, 52, 51, 78, 58, 45, 31, 41, 36, 63, 45, 36, 43, 50, 70, 31, 57, 33, 59, 58, 22, 62, 49, 56, 62, 39, 51, 46, 55, 32, 54, 32, 43, 46, 31, 39, 34, 40, 42, 63, 63, 36, 62, 48, 33, 51, 56, 43, 67, 34, 49, 56, 78, 72, 42, 55, 29, 71, 48, 28, 26, 48, 27, 46, 51, 53, 45 ], "English": [ 36, 64, 48, 15, 44, 17, 28, 39, 54, 32, 27, 39, 25, 14, 27, 17, 28, 39, 44, 29, 19, 34, 37, 29, 34, 41, 41, 14, 24, 25, 39, 24, 10, 46, 12, 21, 14, 39, 21, 27, 15, 37, 33, 20, 42, 19, 33, 27, 28, 28, 20, 29, 32, 27, 29, 42, 20, 17, 67, 33, 37, 13, 43, 28, 34, 19, 22, 43, 55, 40, 27, 14, 19, 25, 26, 25, 27, 31, 25, 15, 36, 31, 26, 28, 18, 46, 19, 85, 30, 20, 16, 16, 31, 31, 39, 13, 19, 29, 38, 44, 41, 26, 22, 27, 32, 26, 69, 42, 20, 24, 23, 26, 18, 29, 26, 22, 38, 35, 45, 32, 33, 20, 29, 16, 27, 46, 22, 13, 16, 15, 18, 45, 29, 38, 42, 16, 32, 22, 31, 63, 31, 34, 27, 32, 22, 20, 45, 46, 23, 21, 29, 24, 31, 34, 18, 31, 26, 33, 21, 35, 27, 24, 23, 20, 39, 25, 21, 20, 36, 23, 26, 33, 44, 55, 40, 31, 45, 36, 42, 22, 21, 45, 17, 32, 37, 23, 32, 28, 20, 23, 34, 33, 15, 22, 11, 9, 19, 26, 25, 29, 17, 35, 38, 18, 21, 24, 30, 32, 14, 19, 35, 23, 45, 45, 21, 24, 23, 35, 32, 36, 34, 37, 23, 37, 26, 36, 18, 21, 25, 23, 35, 36, 33, 28, 30, 24, 34, 33, 26, 18, 24, 30, 36, 35, 19, 31, 26, 33, 19, 40, 22, 34, 40, 14, 36, 24, 36, 26, 22, 31, 27, 26, 17, 30, 18, 25, 29, 20, 27, 32, 24, 26, 38, 38, 19, 33, 25, 19, 29, 36, 32, 31, 21, 31, 39, 46, 33, 25, 24, 22, 38, 18, 24, 16, 27, 15, 28, 33, 32, 28 ], "Hindi": [ 62, 97, 74, 24, 64, 31, 49, 60, 83, 45, 43, 50, 40, 21, 37, 21, 38, 64, 64, 37, 29, 43, 65, 38, 45, 57, 47, 17, 46, 42, 53, 42, 23, 62, 22, 21, 18, 59, 32, 35, 13, 55, 40, 28, 52, 25, 46, 50, 38, 37, 27, 40, 55, 46, 41, 54, 32, 25, 108, 41, 64, 27, 47, 39, 46, 37, 30, 61, 73, 53, 40, 23, 32, 37, 42, 24, 29, 55, 33, 20, 53, 45, 54, 37, 16, 67, 30, 106, 33, 25, 22, 23, 39, 57, 46, 16, 31, 35, 66, 62, 61, 51, 39, 42, 56, 41, 102, 69, 27, 32, 34, 38, 27, 40, 36, 32, 49, 39, 66, 49, 52, 27, 36, 29, 42, 78, 35, 17, 18, 15, 26, 49, 41, 49, 50, 24, 46, 37, 42, 85, 36, 45, 41, 40, 38, 30, 65, 56, 30, 27, 45, 37, 34, 40, 25, 57, 34, 64, 40, 61, 46, 34, 38, 32, 63, 42, 36, 32, 60, 34, 31, 44, 63, 91, 58, 51, 62, 53, 49, 38, 29, 56, 21, 43, 57, 29, 51, 41, 34, 41, 57, 47, 26, 34, 23, 16, 33, 45, 41, 31, 29, 63, 60, 25, 39, 40, 46, 59, 22, 36, 60, 31, 78, 72, 37, 24, 36, 59, 50, 57, 36, 60, 30, 50, 27, 45, 20, 38, 38, 34, 57, 49, 35, 35, 45, 41, 54, 58, 38, 29, 27, 44, 47, 45, 30, 42, 35, 67, 21, 58, 33, 54, 58, 24, 54, 35, 54, 49, 31, 45, 36, 42, 23, 56, 32, 39, 47, 26, 36, 37, 32, 39, 54, 57, 33, 56, 36, 32, 43, 55, 48, 57, 36, 44, 61, 69, 55, 33, 32, 37, 56, 39, 30, 23, 38, 20, 44, 54, 49, 36 ], "Marathi": [ 54, 119, 75, 28, 74, 33, 43, 74, 86, 49, 48, 65, 39, 22, 41, 25, 54, 73, 85, 33, 36, 36, 65, 52, 50, 70, 63, 24, 49, 44, 57, 48, 27, 56, 19, 31, 19, 79, 33, 43, 20, 58, 52, 34, 59, 36, 57, 42, 37, 36, 28, 51, 53, 51, 53, 66, 38, 29, 124, 53, 62, 27, 62, 44, 54, 32, 36, 64, 72, 50, 31, 27, 41, 42, 41, 26, 33, 60, 37, 19, 65, 45, 52, 47, 21, 72, 34, 112, 34, 31, 31, 30, 44, 51, 54, 14, 29, 46, 63, 79, 72, 47, 28, 48, 58, 45, 118, 74, 29, 34, 40, 34, 31, 48, 39, 38, 50, 51, 63, 53, 63, 31, 45, 28, 45, 73, 34, 19, 22, 24, 33, 65, 46, 58, 63, 20, 51, 39, 39, 83, 44, 49, 44, 48, 41, 30, 64, 56, 40, 29, 44, 27, 44, 44, 29, 53, 39, 67, 37, 55, 40, 41, 32, 37, 68, 48, 39, 34, 54, 33, 44, 46, 62, 89, 59, 60, 72, 63, 51, 37, 31, 75, 22, 48, 61, 34, 52, 54, 36, 43, 53, 51, 20, 36, 29, 21, 36, 35, 43, 45, 29, 60, 76, 31, 42, 35, 38, 56, 37, 41, 65, 48, 72, 77, 40, 28, 40, 55, 59, 55, 53, 56, 41, 50, 35, 52, 30, 33, 40, 40, 47, 57, 39, 42, 49, 39, 59, 54, 43, 35, 41, 42, 60, 46, 38, 55, 54, 56, 31, 47, 28, 60, 52, 20, 55, 41, 53, 45, 38, 48, 47, 53, 32, 44, 36, 42, 43, 35, 42, 42, 40, 38, 70, 58, 34, 63, 40, 27, 49, 56, 46, 60, 40, 48, 60, 80, 60, 36, 34, 35, 68, 33, 36, 25, 42, 30, 47, 50, 54, 38 ], "Tamil": [ 61, 118, 100, 30, 90, 46, 49, 66, 81, 58, 53, 72, 45, 23, 40, 28, 50, 111, 80, 44, 32, 39, 74, 61, 63, 77, 58, 31, 48, 41, 47, 40, 25, 68, 27, 42, 20, 85, 39, 44, 22, 73, 46, 26, 68, 27, 61, 41, 36, 52, 36, 52, 51, 49, 66, 64, 38, 28, 109, 50, 62, 37, 69, 55, 50, 31, 35, 70, 97, 60, 31, 24, 48, 52, 37, 39, 27, 50, 37, 22, 75, 60, 55, 42, 15, 76, 31, 118, 48, 28, 27, 23, 41, 49, 95, 16, 33, 45, 59, 81, 68, 40, 31, 46, 64, 45, 132, 73, 33, 35, 47, 43, 47, 50, 38, 31, 47, 39, 78, 48, 60, 34, 45, 31, 42, 81, 41, 22, 28, 20, 28, 73, 59, 59, 60, 25, 55, 46, 38, 92, 38, 49, 31, 46, 42, 31, 66, 63, 37, 23, 50, 38, 45, 44, 30, 52, 42, 74, 39, 61, 48, 51, 27, 39, 60, 42, 42, 38, 68, 47, 39, 55, 74, 112, 62, 61, 72, 61, 62, 39, 33, 68, 32, 52, 61, 40, 55, 48, 40, 60, 64, 52, 34, 32, 32, 25, 36, 36, 36, 43, 34, 61, 64, 22, 33, 37, 49, 56, 28, 42, 56, 55, 89, 81, 42, 29, 38, 51, 59, 53, 47, 61, 40, 64, 35, 55, 30, 45, 40, 51, 77, 68, 40, 46, 52, 38, 62, 57, 39, 31, 44, 43, 55, 45, 28, 60, 40, 82, 29, 60, 38, 58, 76, 24, 62, 45, 72, 56, 37, 50, 53, 51, 24, 66, 30, 41, 44, 34, 31, 41, 38, 42, 58, 61, 38, 62, 46, 29, 51, 58, 56, 62, 44, 47, 63, 78, 64, 37, 40, 31, 66, 46, 39, 24, 31, 23, 49, 48, 51, 31 ], "Telugu": [ 110, 146, 91, 33, 103, 55, 72, 94, 122, 63, 88, 88, 56, 32, 46, 44, 76, 110, 112, 54, 57, 68, 98, 55, 72, 96, 69, 28, 64, 55, 77, 59, 32, 93, 41, 62, 36, 91, 50, 48, 32, 98, 86, 40, 79, 51, 85, 56, 46, 71, 57, 66, 67, 64, 75, 93, 45, 43, 194, 68, 101, 39, 87, 67, 93, 60, 64, 96, 126, 68, 43, 39, 44, 62, 56, 43, 46, 91, 54, 35, 79, 83, 64, 56, 23, 99, 50, 171, 57, 38, 38, 39, 68, 82, 104, 37, 50, 63, 107, 95, 101, 65, 49, 68, 93, 62, 184, 110, 54, 44, 60, 61, 52, 58, 50, 64, 60, 52, 91, 66, 100, 45, 65, 56, 45, 97, 60, 28, 34, 35, 43, 86, 68, 85, 66, 37, 61, 64, 70, 128, 36, 75, 39, 65, 55, 44, 91, 103, 59, 24, 65, 79, 62, 59, 42, 85, 57, 88, 71, 85, 59, 63, 37, 46, 90, 56, 57, 62, 93, 77, 56, 71, 95, 115, 80, 88, 101, 88, 90, 69, 43, 88, 53, 74, 91, 46, 86, 70, 61, 67, 87, 72, 45, 49, 41, 24, 38, 40, 60, 51, 59, 82, 108, 39, 54, 79, 76, 83, 47, 62, 97, 76, 140, 113, 47, 49, 58, 89, 75, 76, 80, 79, 44, 77, 47, 72, 41, 62, 56, 63, 81, 99, 55, 56, 82, 61, 90, 86, 57, 47, 56, 60, 71, 71, 42, 83, 75, 74, 31, 95, 39, 91, 92, 32, 102, 59, 81, 74, 46, 84, 64, 66, 43, 74, 58, 79, 68, 52, 60, 59, 49, 60, 77, 87, 51, 80, 60, 38, 84, 62, 49, 86, 63, 59, 92, 102, 95, 60, 40, 39, 63, 61, 51, 34, 57, 31, 56, 68, 73, 58 ] }, "XLM-R": { "Bengali": [ 53, 94, 77, 25, 63, 32, 45, 66, 87, 47, 41, 59, 41, 16, 39, 24, 44, 69, 63, 38, 30, 43, 62, 33, 49, 62, 49, 19, 46, 31, 54, 40, 22, 51, 21, 33, 16, 64, 28, 36, 17, 56, 52, 26, 45, 31, 52, 34, 33, 39, 25, 42, 47, 39, 49, 57, 30, 28, 116, 48, 60, 29, 64, 58, 56, 31, 36, 61, 73, 48, 21, 18, 26, 33, 42, 33, 38, 54, 36, 13, 62, 38, 51, 27, 25, 59, 30, 99, 39, 30, 22, 21, 35, 56, 66, 14, 28, 30, 52, 66, 48, 33, 37, 44, 56, 43, 101, 73, 30, 34, 38, 34, 23, 36, 33, 31, 47, 50, 59, 41, 48, 28, 31, 33, 42, 72, 39, 14, 18, 21, 27, 49, 43, 57, 76, 20, 46, 34, 38, 68, 39, 54, 32, 44, 42, 24, 56, 44, 37, 26, 48, 33, 33, 42, 33, 44, 31, 92, 34, 68, 32, 34, 26, 24, 74, 39, 36, 33, 62, 34, 38, 53, 57, 70, 66, 53, 65, 51, 56, 43, 33, 57, 17, 45, 61, 30, 53, 46, 29, 41, 64, 45, 20, 35, 24, 16, 29, 33, 38, 38, 37, 50, 58, 23, 39, 39, 47, 59, 23, 31, 56, 33, 68, 69, 31, 30, 25, 48, 55, 48, 43, 55, 33, 48, 27, 48, 24, 41, 30, 38, 44, 60, 33, 31, 51, 45, 61, 51, 35, 28, 33, 36, 57, 43, 22, 39, 45, 58, 33, 51, 31, 53, 44, 19, 51, 41, 52, 37, 25, 43, 38, 46, 30, 47, 31, 33, 34, 22, 35, 28, 36, 40, 47, 56, 27, 49, 45, 30, 46, 53, 42, 59, 35, 39, 52, 69, 64, 38, 49, 28, 61, 40, 25, 19, 41, 23, 40, 49, 50, 37 ], "English": [ 36, 65, 50, 17, 44, 20, 29, 40, 59, 35, 32, 45, 26, 15, 27, 13, 28, 39, 42, 29, 24, 35, 39, 27, 35, 43, 41, 16, 28, 29, 41, 27, 11, 35, 15, 23, 14, 41, 20, 25, 15, 45, 38, 13, 43, 23, 34, 29, 24, 31, 20, 32, 31, 29, 31, 45, 21, 19, 72, 33, 38, 16, 47, 28, 37, 22, 22, 44, 58, 39, 25, 15, 21, 28, 25, 26, 29, 33, 27, 16, 39, 31, 30, 30, 18, 44, 21, 75, 30, 21, 17, 18, 33, 35, 44, 15, 19, 26, 39, 46, 40, 26, 24, 26, 35, 29, 71, 47, 21, 24, 27, 29, 20, 31, 28, 23, 40, 36, 48, 35, 36, 22, 30, 20, 28, 46, 24, 15, 16, 15, 19, 43, 25, 41, 45, 17, 35, 23, 35, 55, 32, 33, 23, 30, 25, 21, 47, 48, 25, 23, 32, 26, 35, 34, 20, 41, 24, 43, 21, 36, 29, 26, 25, 25, 38, 27, 23, 23, 41, 27, 26, 38, 46, 59, 43, 33, 44, 36, 43, 26, 22, 43, 19, 34, 39, 24, 35, 29, 23, 24, 41, 31, 17, 24, 14, 10, 20, 29, 28, 31, 19, 36, 40, 18, 25, 27, 32, 35, 17, 23, 36, 25, 46, 41, 20, 24, 24, 37, 36, 34, 36, 38, 23, 38, 27, 40, 18, 22, 26, 25, 34, 39, 28, 26, 33, 28, 33, 35, 27, 18, 25, 30, 39, 37, 19, 33, 31, 37, 20, 43, 24, 37, 34, 14, 39, 26, 38, 29, 23, 33, 30, 28, 19, 32, 19, 26, 31, 20, 24, 33, 27, 28, 44, 38, 21, 31, 26, 20, 29, 38, 31, 33, 23, 33, 41, 49, 36, 25, 28, 25, 38, 21, 25, 19, 30, 15, 29, 38, 36, 30 ], "Hindi": [ 52, 79, 64, 21, 57, 31, 44, 57, 77, 42, 36, 45, 41, 21, 32, 16, 35, 59, 56, 36, 25, 41, 54, 29, 38, 50, 42, 15, 38, 29, 47, 40, 23, 48, 21, 18, 16, 56, 28, 31, 11, 56, 36, 19, 48, 22, 42, 44, 26, 26, 18, 38, 46, 47, 35, 52, 29, 23, 93, 35, 52, 21, 44, 39, 44, 32, 32, 56, 65, 40, 33, 24, 28, 31, 35, 20, 27, 57, 31, 20, 49, 34, 54, 34, 15, 62, 30, 85, 29, 23, 22, 22, 40, 49, 45, 15, 30, 31, 53, 53, 53, 42, 31, 32, 53, 39, 96, 62, 27, 29, 27, 33, 24, 40, 32, 31, 48, 34, 62, 45, 45, 25, 32, 23, 39, 71, 31, 12, 15, 15, 23, 35, 32, 44, 45, 22, 42, 35, 37, 61, 34, 38, 34, 38, 33, 23, 57, 44, 28, 25, 41, 36, 30, 30, 23, 46, 28, 51, 34, 58, 43, 24, 35, 28, 58, 35, 32, 28, 53, 31, 26, 40, 54, 76, 53, 45, 59, 44, 43, 32, 26, 47, 22, 37, 52, 31, 48, 37, 32, 39, 51, 38, 21, 30, 19, 14, 30, 43, 37, 31, 29, 58, 60, 24, 33, 37, 42, 47, 17, 27, 49, 27, 77, 56, 31, 23, 33, 51, 48, 49, 32, 56, 28, 46, 24, 43, 19, 38, 37, 35, 48, 38, 30, 26, 43, 33, 46, 52, 33, 27, 27, 39, 46, 41, 32, 42, 33, 55, 22, 56, 30, 55, 46, 21, 47, 33, 48, 41, 24, 42, 32, 37, 19, 49, 28, 37, 46, 24, 29, 34, 31, 39, 49, 52, 33, 49, 31, 26, 42, 43, 41, 50, 32, 34, 58, 65, 51, 26, 30, 32, 48, 31, 29, 18, 39, 21, 41, 43, 45, 30 ], "Marathi": [ 40, 91, 58, 23, 64, 31, 37, 58, 71, 41, 37, 50, 33, 15, 33, 18, 40, 61, 63, 26, 29, 31, 48, 34, 46, 59, 46, 20, 38, 32, 41, 39, 21, 39, 17, 28, 15, 63, 24, 36, 14, 54, 40, 22, 45, 24, 50, 32, 27, 28, 19, 40, 40, 41, 43, 54, 31, 25, 97, 45, 46, 21, 53, 42, 51, 24, 29, 53, 64, 34, 22, 19, 30, 33, 29, 19, 29, 45, 29, 18, 55, 32, 46, 34, 17, 53, 28, 89, 32, 23, 23, 24, 31, 42, 52, 14, 24, 32, 51, 61, 48, 38, 19, 38, 50, 35, 99, 58, 25, 24, 33, 28, 23, 43, 31, 27, 43, 37, 51, 44, 44, 25, 35, 21, 31, 65, 30, 15, 16, 17, 25, 48, 32, 48, 47, 15, 44, 31, 39, 59, 38, 38, 34, 39, 34, 23, 44, 45, 29, 22, 41, 23, 33, 27, 20, 35, 27, 56, 28, 49, 34, 34, 26, 26, 52, 35, 32, 29, 47, 27, 36, 39, 40, 70, 47, 42, 54, 42, 39, 25, 27, 58, 21, 40, 48, 30, 45, 38, 26, 32, 40, 40, 14, 24, 23, 17, 26, 27, 34, 35, 22, 44, 59, 29, 32, 27, 28, 46, 29, 30, 52, 41, 56, 64, 32, 26, 32, 40, 48, 42, 35, 46, 32, 39, 30, 42, 19, 27, 28, 29, 39, 42, 31, 27, 42, 32, 41, 36, 30, 29, 31, 29, 57, 33, 27, 40, 39, 46, 29, 42, 24, 45, 39, 15, 47, 33, 44, 31, 26, 36, 33, 38, 23, 43, 30, 31, 33, 31, 29, 30, 31, 31, 59, 43, 27, 47, 34, 25, 41, 37, 34, 49, 31, 37, 48, 62, 48, 24, 27, 28, 47, 21, 31, 22, 32, 21, 41, 45, 40, 28 ], "Tamil": [ 48, 97, 81, 24, 73, 38, 43, 56, 68, 47, 47, 61, 37, 21, 34, 18, 30, 81, 62, 34, 28, 37, 71, 45, 55, 74, 50, 25, 44, 40, 42, 39, 21, 54, 26, 37, 15, 72, 30, 34, 15, 59, 40, 18, 51, 21, 58, 35, 23, 44, 31, 45, 37, 40, 57, 58, 33, 27, 95, 39, 51, 25, 62, 58, 44, 22, 34, 56, 77, 43, 22, 21, 35, 41, 30, 33, 26, 42, 33, 21, 66, 45, 49, 39, 12, 64, 27, 91, 44, 23, 24, 21, 35, 42, 79, 16, 25, 34, 48, 71, 59, 32, 25, 35, 58, 42, 108, 68, 31, 27, 44, 35, 31, 38, 31, 27, 41, 38, 69, 39, 54, 26, 36, 25, 35, 78, 37, 19, 26, 18, 20, 59, 42, 53, 48, 20, 44, 37, 33, 68, 32, 38, 22, 37, 36, 27, 50, 52, 33, 22, 48, 28, 42, 33, 25, 41, 33, 66, 37, 49, 37, 43, 23, 33, 57, 39, 30, 30, 64, 41, 31, 49, 63, 96, 50, 48, 60, 49, 56, 34, 27, 54, 24, 47, 47, 33, 47, 43, 32, 49, 53, 37, 28, 26, 24, 23, 31, 29, 31, 37, 30, 48, 58, 21, 29, 30, 45, 48, 23, 36, 40, 39, 67, 75, 37, 29, 32, 41, 48, 43, 36, 55, 35, 53, 33, 46, 24, 38, 32, 42, 65, 58, 33, 41, 45, 32, 44, 47, 35, 31, 41, 36, 45, 41, 24, 49, 32, 70, 30, 57, 34, 54, 58, 20, 50, 34, 55, 50, 32, 40, 46, 42, 22, 53, 25, 30, 34, 25, 26, 35, 31, 35, 46, 51, 33, 50, 38, 28, 39, 48, 44, 53, 41, 35, 58, 69, 60, 33, 33, 26, 53, 38, 32, 19, 24, 20, 44, 44, 45, 25 ], "Telugu": [ 53, 80, 52, 20, 66, 28, 42, 56, 69, 35, 45, 54, 34, 15, 24, 21, 38, 65, 59, 32, 25, 34, 52, 27, 45, 64, 36, 18, 38, 36, 46, 33, 17, 51, 23, 39, 21, 47, 24, 30, 13, 61, 50, 16, 38, 26, 60, 34, 29, 45, 35, 39, 36, 39, 41, 47, 35, 34, 123, 40, 57, 25, 54, 58, 65, 31, 34, 56, 69, 34, 23, 21, 23, 28, 33, 30, 32, 53, 32, 18, 53, 43, 48, 27, 12, 67, 24, 97, 35, 19, 21, 26, 41, 49, 64, 22, 30, 28, 62, 66, 60, 41, 27, 32, 56, 38, 111, 68, 28, 24, 38, 43, 25, 29, 34, 33, 44, 37, 60, 50, 59, 24, 33, 39, 28, 56, 32, 16, 17, 26, 26, 55, 49, 51, 34, 17, 34, 38, 59, 72, 31, 40, 28, 43, 32, 37, 49, 52, 42, 18, 42, 49, 39, 33, 19, 40, 27, 48, 39, 50, 32, 38, 17, 20, 60, 32, 36, 48, 60, 36, 35, 53, 54, 72, 48, 49, 52, 46, 54, 43, 28, 49, 23, 45, 55, 30, 50, 46, 26, 37, 50, 38, 25, 28, 24, 16, 25, 23, 34, 30, 36, 49, 59, 24, 35, 49, 42, 49, 27, 36, 45, 45, 57, 66, 27, 30, 26, 50, 55, 40, 38, 49, 27, 46, 22, 47, 31, 37, 31, 38, 48, 67, 32, 35, 49, 37, 48, 42, 34, 24, 29, 28, 56, 35, 24, 45, 48, 51, 25, 50, 28, 55, 52, 19, 56, 34, 48, 41, 29, 49, 32, 33, 31, 53, 31, 43, 43, 30, 29, 30, 27, 36, 40, 46, 30, 48, 39, 26, 56, 35, 32, 49, 28, 35, 61, 58, 62, 31, 27, 23, 35, 25, 35, 19, 37, 23, 37, 41, 45, 40 ] }, "mBERT": { "Bengali": [ 70, 119, 92, 34, 85, 41, 48, 78, 97, 51, 66, 67, 49, 19, 50, 35, 52, 92, 75, 48, 42, 51, 81, 45, 73, 84, 70, 27, 49, 55, 57, 52, 30, 62, 33, 40, 31, 83, 36, 30, 18, 80, 63, 31, 59, 37, 71, 58, 35, 58, 48, 53, 57, 48, 65, 77, 38, 34, 138, 57, 71, 37, 87, 70, 65, 41, 42, 69, 98, 61, 32, 25, 40, 45, 46, 36, 40, 64, 50, 16, 71, 62, 57, 42, 34, 73, 38, 116, 54, 30, 31, 29, 60, 76, 85, 18, 33, 41, 61, 81, 62, 44, 44, 54, 74, 58, 132, 76, 36, 38, 52, 36, 24, 53, 42, 39, 53, 53, 71, 54, 60, 34, 44, 37, 49, 81, 49, 20, 24, 27, 32, 62, 50, 76, 99, 24, 67, 39, 49, 83, 44, 67, 41, 51, 56, 30, 62, 60, 52, 33, 51, 39, 51, 41, 42, 69, 35, 109, 53, 74, 39, 60, 37, 39, 84, 56, 42, 40, 83, 52, 57, 71, 74, 92, 85, 72, 73, 76, 79, 41, 40, 60, 28, 52, 68, 42, 44, 64, 47, 56, 74, 67, 27, 40, 28, 21, 40, 46, 53, 41, 42, 65, 77, 25, 42, 51, 60, 72, 40, 49, 79, 53, 98, 93, 35, 31, 41, 59, 70, 50, 60, 69, 44, 69, 34, 70, 31, 57, 39, 65, 58, 80, 33, 44, 53, 63, 69, 71, 59, 33, 39, 46, 65, 60, 34, 57, 56, 75, 38, 48, 44, 60, 53, 25, 76, 57, 65, 78, 46, 56, 53, 63, 34, 55, 38, 35, 55, 28, 47, 36, 52, 51, 69, 71, 35, 58, 52, 40, 48, 67, 48, 72, 39, 48, 70, 94, 74, 48, 46, 32, 77, 54, 33, 28, 54, 30, 52, 49, 69, 51 ], "English": [ 35, 57, 50, 13, 40, 21, 26, 38, 54, 33, 32, 40, 26, 14, 27, 14, 28, 37, 37, 30, 20, 34, 37, 27, 33, 41, 40, 15, 22, 23, 35, 25, 10, 36, 11, 21, 14, 41, 18, 27, 15, 42, 33, 15, 41, 16, 35, 28, 25, 26, 21, 28, 26, 26, 28, 45, 18, 18, 65, 30, 37, 13, 45, 27, 35, 16, 20, 42, 55, 36, 21, 13, 17, 23, 25, 25, 28, 33, 28, 16, 37, 31, 34, 29, 18, 43, 18, 72, 28, 21, 17, 15, 30, 34, 35, 13, 19, 25, 38, 48, 35, 26, 21, 25, 31, 26, 65, 44, 23, 26, 26, 28, 18, 31, 24, 19, 38, 34, 46, 33, 34, 21, 28, 19, 27, 47, 22, 14, 16, 14, 17, 37, 23, 38, 43, 17, 32, 21, 29, 49, 30, 32, 22, 30, 21, 17, 40, 44, 23, 22, 29, 25, 33, 27, 18, 34, 22, 35, 22, 36, 29, 26, 25, 21, 38, 26, 19, 22, 37, 24, 27, 40, 42, 59, 40, 32, 48, 35, 43, 26, 20, 40, 19, 31, 36, 25, 32, 26, 20, 21, 36, 32, 15, 22, 11, 9, 19, 24, 26, 30, 17, 35, 41, 18, 21, 25, 32, 34, 15, 19, 34, 24, 49, 43, 22, 23, 23, 35, 32, 31, 35, 35, 21, 37, 25, 38, 18, 22, 25, 23, 29, 36, 28, 24, 30, 24, 31, 33, 26, 19, 23, 29, 42, 35, 19, 36, 28, 30, 20, 41, 23, 34, 32, 14, 37, 25, 38, 26, 22, 30, 27, 27, 19, 26, 18, 26, 30, 20, 24, 32, 24, 25, 42, 40, 22, 32, 26, 18, 28, 37, 31, 32, 21, 30, 41, 50, 37, 24, 27, 22, 39, 19, 24, 17, 28, 15, 29, 33, 32, 28 ], "Hindi": [ 80, 107, 94, 29, 73, 41, 59, 69, 101, 58, 58, 58, 47, 24, 41, 24, 54, 74, 73, 44, 38, 51, 81, 47, 63, 70, 63, 23, 56, 50, 58, 52, 31, 59, 28, 26, 26, 87, 34, 40, 16, 72, 40, 24, 59, 32, 54, 55, 37, 44, 35, 56, 59, 59, 51, 69, 38, 30, 126, 50, 79, 30, 57, 44, 51, 37, 34, 76, 93, 55, 41, 29, 42, 40, 41, 29, 32, 65, 40, 25, 65, 49, 66, 44, 23, 87, 39, 113, 45, 26, 29, 32, 46, 66, 61, 19, 41, 42, 76, 84, 67, 63, 39, 50, 65, 53, 127, 76, 31, 41, 43, 45, 31, 49, 38, 37, 55, 43, 76, 56, 63, 33, 42, 31, 48, 91, 39, 22, 22, 18, 29, 53, 37, 54, 62, 28, 54, 46, 41, 80, 45, 39, 44, 48, 43, 36, 72, 64, 35, 33, 63, 43, 44, 36, 28, 68, 38, 75, 51, 76, 59, 47, 41, 37, 78, 49, 39, 35, 71, 44, 40, 55, 79, 96, 71, 62, 69, 67, 67, 49, 31, 64, 29, 48, 67, 35, 61, 50, 43, 59, 71, 59, 25, 37, 24, 20, 41, 49, 47, 37, 37, 76, 78, 29, 46, 46, 51, 70, 33, 39, 75, 43, 99, 89, 38, 27, 41, 63, 58, 55, 45, 72, 35, 54, 32, 53, 28, 52, 43, 45, 60, 57, 34, 37, 50, 46, 56, 70, 46, 38, 34, 53, 58, 49, 35, 52, 44, 79, 23, 64, 36, 60, 60, 31, 74, 44, 69, 55, 40, 54, 46, 51, 26, 58, 39, 46, 62, 27, 40, 39, 44, 49, 69, 61, 38, 58, 36, 41, 51, 65, 65, 65, 42, 42, 73, 87, 59, 32, 41, 48, 63, 47, 34, 25, 49, 25, 57, 60, 63, 40 ], "Marathi": [ 62, 146, 99, 36, 80, 40, 56, 82, 101, 58, 58, 76, 47, 23, 45, 30, 69, 90, 96, 44, 46, 38, 76, 63, 65, 94, 75, 35, 54, 58, 66, 56, 37, 59, 27, 41, 30, 105, 36, 50, 23, 75, 58, 28, 68, 40, 66, 49, 37, 43, 39, 63, 60, 64, 67, 76, 40, 32, 143, 58, 71, 30, 74, 50, 71, 38, 42, 74, 79, 52, 34, 31, 48, 48, 47, 28, 37, 69, 55, 25, 78, 50, 64, 50, 28, 87, 44, 129, 49, 36, 33, 36, 55, 56, 72, 18, 37, 49, 78, 101, 80, 62, 34, 59, 70, 60, 158, 93, 34, 37, 53, 41, 37, 59, 40, 48, 58, 50, 78, 58, 80, 39, 52, 32, 56, 82, 38, 23, 29, 30, 38, 73, 47, 61, 79, 26, 69, 46, 49, 88, 50, 52, 50, 56, 48, 37, 68, 66, 47, 35, 59, 34, 50, 43, 38, 69, 37, 80, 43, 78, 46, 50, 36, 48, 92, 57, 44, 42, 64, 45, 49, 59, 74, 101, 68, 67, 82, 88, 69, 44, 37, 84, 31, 63, 72, 41, 62, 64, 45, 53, 61, 66, 27, 43, 30, 26, 51, 44, 50, 52, 33, 71, 98, 35, 50, 41, 50, 66, 47, 47, 78, 66, 94, 98, 43, 37, 50, 67, 68, 54, 64, 66, 50, 59, 43, 58, 35, 48, 49, 50, 56, 63, 39, 40, 61, 51, 57, 61, 50, 47, 50, 54, 73, 64, 45, 61, 61, 68, 35, 56, 31, 64, 64, 23, 69, 55, 71, 54, 46, 57, 52, 68, 31, 50, 49, 47, 60, 35, 49, 46, 55, 49, 91, 70, 37, 69, 46, 30, 60, 69, 50, 69, 43, 51, 70, 95, 64, 42, 34, 47, 81, 46, 41, 31, 50, 36, 51, 62, 70, 53 ], "Tamil": [ 79, 144, 124, 34, 98, 57, 68, 95, 103, 64, 70, 95, 54, 30, 57, 28, 63, 128, 98, 55, 41, 50, 94, 61, 81, 94, 80, 35, 57, 48, 59, 53, 35, 81, 31, 54, 28, 121, 43, 49, 28, 93, 61, 25, 78, 40, 84, 56, 35, 61, 47, 62, 62, 69, 101, 84, 44, 39, 130, 65, 70, 41, 90, 75, 69, 35, 46, 79, 116, 69, 36, 27, 51, 65, 48, 44, 38, 68, 56, 31, 94, 81, 70, 61, 18, 87, 40, 127, 58, 35, 39, 30, 57, 64, 99, 16, 46, 51, 80, 123, 87, 44, 44, 53, 84, 67, 157, 89, 44, 36, 62, 47, 61, 68, 37, 38, 51, 44, 93, 57, 83, 43, 52, 41, 51, 96, 57, 26, 34, 31, 35, 87, 57, 74, 77, 36, 68, 64, 49, 91, 37, 55, 29, 51, 50, 38, 67, 78, 55, 25, 58, 43, 59, 48, 36, 70, 40, 93, 47, 71, 63, 64, 38, 56, 79, 60, 48, 42, 78, 59, 47, 65, 85, 128, 74, 74, 82, 86, 74, 49, 38, 75, 37, 60, 72, 47, 65, 69, 48, 74, 83, 63, 44, 40, 32, 28, 40, 43, 42, 48, 45, 75, 81, 29, 43, 50, 70, 67, 35, 57, 67, 68, 96, 98, 50, 36, 52, 66, 74, 54, 68, 75, 47, 83, 48, 75, 37, 68, 41, 69, 87, 83, 39, 53, 60, 50, 61, 72, 50, 44, 51, 48, 72, 65, 32, 86, 50, 97, 39, 75, 50, 68, 82, 29, 76, 56, 92, 75, 51, 55, 79, 78, 30, 74, 48, 60, 54, 39, 45, 55, 49, 57, 88, 76, 39, 68, 49, 36, 58, 86, 70, 82, 47, 54, 75, 106, 81, 48, 56, 42, 71, 52, 41, 27, 41, 33, 63, 58, 71, 43 ], "Telugu": [ 101, 127, 83, 34, 83, 47, 74, 90, 109, 49, 81, 74, 52, 32, 44, 38, 74, 98, 106, 51, 41, 56, 81, 49, 63, 107, 59, 28, 57, 56, 65, 51, 33, 77, 36, 56, 28, 72, 40, 47, 32, 92, 76, 28, 63, 40, 79, 60, 49, 64, 49, 57, 58, 53, 74, 84, 42, 42, 172, 59, 89, 34, 82, 73, 99, 47, 51, 89, 113, 59, 39, 33, 43, 51, 47, 37, 41, 72, 49, 28, 84, 66, 71, 53, 21, 90, 39, 156, 53, 33, 31, 32, 62, 80, 97, 32, 50, 47, 91, 92, 94, 64, 45, 62, 82, 55, 169, 101, 43, 38, 60, 48, 46, 52, 47, 51, 59, 48, 84, 62, 87, 36, 53, 36, 36, 82, 56, 21, 32, 16, 42, 81, 63, 81, 60, 29, 49, 54, 71, 92, 33, 65, 36, 61, 50, 21, 80, 80, 30, 21, 64, 66, 50, 47, 35, 72, 50, 71, 75, 91, 51, 49, 31, 30, 88, 48, 56, 39, 79, 69, 52, 68, 81, 93, 76, 79, 85, 88, 82, 55, 43, 76, 43, 63, 76, 40, 62, 64, 44, 75, 92, 66, 45, 52, 35, 21, 29, 34, 53, 45, 54, 73, 96, 38, 49, 68, 70, 85, 50, 60, 76, 76, 108, 107, 42, 38, 63, 81, 76, 62, 89, 76, 33, 77, 42, 64, 39, 61, 46, 69, 61, 70, 43, 58, 60, 49, 65, 75, 55, 43, 49, 48, 62, 58, 38, 81, 72, 57, 30, 64, 38, 71, 78, 31, 94, 54, 71, 72, 41, 72, 61, 68, 42, 62, 51, 74, 69, 38, 52, 49, 45, 52, 75, 69, 50, 63, 49, 36, 50, 60, 49, 91, 47, 52, 81, 93, 85, 50, 39, 43, 58, 53, 46, 33, 57, 29, 60, 61, 72, 51 ] } }, "language_codes": { "Bengali": "ben_Beng", "English": "eng_Latn", "Hindi": "hin_Deva", "Marathi": "mar_Deva", "Tamil": "tam_Taml", "Telugu": "tel_Telu" }, "machine": "x86_64", "model_order": [ "GPT-2", "mBERT", "XLM-R", "Qwen2.5", "UMT5" ], "pieces": { "GPT-2": { "English": { "ids": [ 7594, 617, 584, 6154, 11 ], "tokens": [ "Like", "Ġsome", "Ġother", "Ġexperts", "," ], "total_tokens": 31 }, "Hindi": { "ids": [ 11976, 243, 24231, 223, 11976 ], "tokens": [ "à¤", "ķ", "à¥", "ģ", "à¤" ], "total_tokens": 308 } }, "Qwen2.5": { "English": { "ids": [ 12949, 1045, 1008, 11647, 11 ], "tokens": [ "Like", "Ġsome", "Ġother", "Ġexperts", "," ], "total_tokens": 32 }, "Hindi": { "ids": [ 64704, 72653, 147877, 14925, 227 ], "tokens": [ "à¤ķ", "à¥ģ", "à¤Ľ", "Ġà¤", "ħ" ], "total_tokens": 182 } }, "mBERT": { "English": { "ids": [ 15105, 11152, 10684, 47306, 117 ], "tokens": [ "Like", "some", "other", "experts", "," ], "total_tokens": 35 }, "Hindi": { "ids": [ 22022, 22197, 37100, 80452, 11497 ], "tokens": [ "कुछ", "अन्य", "विशेष", "##ज्ञ", "##ों" ], "total_tokens": 80 } } }, "platform": "macOS-26.5.2-x86_64-i386-64bit", "population_n": 1012, "python": "3.9.10", "run_id": "28-paper4-20261010-02", "sample_indices_zero_based": [ 2, 3, 15, 21, 23, 25, 26, 27, 31, 32, 33, 36, 37, 39, 44, 50, 53, 55, 56, 60, 61, 62, 63, 68, 69, 72, 74, 78, 80, 82, 83, 88, 93, 102, 103, 105, 106, 107, 109, 110, 111, 114, 116, 118, 121, 122, 125, 128, 137, 146, 149, 150, 153, 154, 155, 156, 159, 160, 161, 167, 168, 172, 184, 188, 189, 195, 199, 200, 203, 204, 206, 212, 221, 226, 228, 231, 235, 237, 242, 245, 248, 255, 259, 260, 262, 265, 273, 275, 279, 281, 284, 285, 287, 294, 295, 298, 307, 315, 320, 321, 323, 324, 327, 332, 335, 339, 340, 343, 344, 348, 349, 351, 354, 357, 363, 364, 367, 368, 369, 370, 373, 374, 376, 379, 384, 385, 391, 393, 394, 395, 399, 403, 408, 412, 415, 416, 417, 420, 422, 426, 427, 430, 432, 433, 438, 442, 451, 454, 456, 463, 467, 469, 477, 480, 483, 485, 494, 497, 508, 511, 512, 515, 516, 518, 520, 522, 525, 529, 531, 534, 539, 542, 544, 556, 559, 562, 564, 565, 566, 580, 582, 585, 594, 596, 597, 600, 601, 603, 611, 614, 615, 616, 620, 622, 632, 634, 640, 644, 648, 653, 654, 656, 658, 662, 663, 669, 670, 673, 674, 676, 682, 685, 686, 689, 694, 702, 703, 705, 709, 713, 714, 716, 720, 731, 738, 745, 750, 753, 755, 758, 764, 766, 768, 771, 774, 777, 785, 793, 803, 804, 805, 806, 811, 818, 821, 822, 825, 828, 837, 838, 842, 847, 852, 855, 857, 859, 860, 877, 879, 889, 895, 896, 902, 904, 912, 914, 915, 920, 924, 927, 930, 931, 932, 934, 937, 942, 945, 946, 950, 951, 953, 955, 956, 963, 965, 967, 970, 976, 977, 980, 983, 984, 985, 986, 993, 994, 1000, 1004, 1005, 1007 ], "sample_n": 300, "seed": 28004, "stats": { "GPT-2": { "Bengali": { "context_4096_sentences": 15, "ratio_ci95": [ 9.3758, 9.7766 ], "ratio_to_english": 9.5781, "tokens_per_character": 2.0033, "tokens_per_sentence": 258.6733 }, "English": { "context_4096_sentences": 151, "ratio_ci95": [ 1.0, 1.0 ], "ratio_to_english": 1.0, "tokens_per_character": 0.2056, "tokens_per_sentence": 27.0067 }, "Hindi": { "context_4096_sentences": 20, "ratio_ci95": [ 7.2617, 7.5918 ], "ratio_to_english": 7.4269, "tokens_per_character": 1.5333, "tokens_per_sentence": 200.5767 }, "Marathi": { "context_4096_sentences": 19, "ratio_ci95": [ 7.5609, 7.8723 ], "ratio_to_english": 7.7183, "tokens_per_character": 1.5961, "tokens_per_sentence": 208.4467 }, "Tamil": { "context_4096_sentences": 9, "ratio_ci95": [ 15.2287, 15.9183 ], "ratio_to_english": 15.5755, "tokens_per_character": 2.7309, "tokens_per_sentence": 420.6433 }, "Telugu": { "context_4096_sentences": 11, "ratio_ci95": [ 12.5528, 13.1202 ], "ratio_to_english": 12.8383, "tokens_per_character": 2.6419, "tokens_per_sentence": 346.72 } }, "Qwen2.5": { "Bengali": { "context_4096_sentences": 29, "ratio_ci95": [ 4.9549, 5.176 ], "ratio_to_english": 5.0677, "tokens_per_character": 1.0747, "tokens_per_sentence": 138.77 }, "English": { "context_4096_sentences": 149, "ratio_ci95": [ 1.0, 1.0 ], "ratio_to_english": 1.0, "tokens_per_character": 0.2085, "tokens_per_sentence": 27.3833 }, "Hindi": { "context_4096_sentences": 33, "ratio_ci95": [ 4.3422, 4.5439 ], "ratio_to_english": 4.4441, "tokens_per_character": 0.9303, "tokens_per_sentence": 121.6933 }, "Marathi": { "context_4096_sentences": 32, "ratio_ci95": [ 4.4672, 4.6504 ], "ratio_to_english": 4.5579, "tokens_per_character": 0.9557, "tokens_per_sentence": 124.81 }, "Tamil": { "context_4096_sentences": 24, "ratio_ci95": [ 6.0254, 6.2909 ], "ratio_to_english": 6.1621, "tokens_per_character": 1.0955, "tokens_per_sentence": 168.74 }, "Telugu": { "context_4096_sentences": 21, "ratio_ci95": [ 6.8123, 7.1354 ], "ratio_to_english": 6.9744, "tokens_per_character": 1.4553, "tokens_per_sentence": 190.9833 } }, "UMT5": { "Bengali": { "context_4096_sentences": 83, "ratio_ci95": [ 1.6454, 1.7049 ], "ratio_to_english": 1.6755, "tokens_per_character": 0.3777, "tokens_per_sentence": 48.7733 }, "English": { "context_4096_sentences": 140, "ratio_ci95": [ 1.0, 1.0 ], "ratio_to_english": 1.0, "tokens_per_character": 0.2217, "tokens_per_sentence": 29.11 }, "Hindi": { "context_4096_sentences": 95, "ratio_ci95": [ 1.4506, 1.4998 ], "ratio_to_english": 1.4749, "tokens_per_character": 0.3282, "tokens_per_sentence": 42.9333 }, "Marathi": { "context_4096_sentences": 87, "ratio_ci95": [ 1.5796, 1.6299 ], "ratio_to_english": 1.6045, "tokens_per_character": 0.3576, "tokens_per_sentence": 46.7067 }, "Tamil": { "context_4096_sentences": 83, "ratio_ci95": [ 1.6608, 1.7275 ], "ratio_to_english": 1.6939, "tokens_per_character": 0.3201, "tokens_per_sentence": 49.31 }, "Telugu": { "context_4096_sentences": 60, "ratio_ci95": [ 2.2834, 2.3802 ], "ratio_to_english": 2.3324, "tokens_per_character": 0.5174, "tokens_per_sentence": 67.8967 } }, "XLM-R": { "Bengali": { "context_4096_sentences": 96, "ratio_ci95": [ 1.3664, 1.416 ], "ratio_to_english": 1.3907, "tokens_per_character": 0.3291, "tokens_per_sentence": 42.5 }, "English": { "context_4096_sentences": 134, "ratio_ci95": [ 1.0, 1.0 ], "ratio_to_english": 1.0, "tokens_per_character": 0.2327, "tokens_per_sentence": 30.56 }, "Hindi": { "context_4096_sentences": 107, "ratio_ci95": [ 1.228, 1.2696 ], "ratio_to_english": 1.2486, "tokens_per_character": 0.2917, "tokens_per_sentence": 38.1567 }, "Marathi": { "context_4096_sentences": 111, "ratio_ci95": [ 1.1778, 1.2185 ], "ratio_to_english": 1.1984, "tokens_per_character": 0.2804, "tokens_per_sentence": 36.6233 }, "Tamil": { "context_4096_sentences": 99, "ratio_ci95": [ 1.3262, 1.3787 ], "ratio_to_english": 1.3511, "tokens_per_character": 0.2681, "tokens_per_sentence": 41.29 }, "Telugu": { "context_4096_sentences": 103, "ratio_ci95": [ 1.2717, 1.3303 ], "ratio_to_english": 1.3007, "tokens_per_character": 0.3029, "tokens_per_sentence": 39.75 } }, "mBERT": { "Bengali": { "context_4096_sentences": 75, "ratio_ci95": [ 1.8429, 1.9187 ], "ratio_to_english": 1.8822, "tokens_per_character": 0.4199, "tokens_per_sentence": 54.2267 }, "English": { "context_4096_sentences": 142, "ratio_ci95": [ 1.0, 1.0 ], "ratio_to_english": 1.0, "tokens_per_character": 0.2194, "tokens_per_sentence": 28.81 }, "Hindi": { "context_4096_sentences": 80, "ratio_ci95": [ 1.7284, 1.7935 ], "ratio_to_english": 1.7614, "tokens_per_character": 0.3879, "tokens_per_sentence": 50.7467 }, "Marathi": { "context_4096_sentences": 73, "ratio_ci95": [ 1.9015, 1.974 ], "ratio_to_english": 1.9361, "tokens_per_character": 0.4271, "tokens_per_sentence": 55.78 }, "Tamil": { "context_4096_sentences": 67, "ratio_ci95": [ 2.0746, 2.1635 ], "ratio_to_english": 2.1171, "tokens_per_character": 0.396, "tokens_per_sentence": 60.9933 }, "Telugu": { "context_4096_sentences": 67, "ratio_ci95": [ 2.049, 2.1413 ], "ratio_to_english": 2.0957, "tokens_per_character": 0.4601, "tokens_per_sentence": 60.3767 } } }, "tokenizer_config": { "GPT-2": { "family": "byte BPE", "padding": null, "truncation": null, "unknown_id": null, "unknown_token": null }, "Qwen2.5": { "family": "byte BPE", "padding": null, "truncation": null, "unknown_id": null, "unknown_token": null }, "UMT5": { "family": "SentencePiece Unigram", "padding": null, "truncation": null, "unknown_id": 3, "unknown_token": "" }, "XLM-R": { "family": "SentencePiece Unigram", "padding": null, "truncation": null, "unknown_id": 3, "unknown_token": "" }, "mBERT": { "family": "WordPiece", "padding": null, "truncation": null, "unknown_id": 100, "unknown_token": "[UNK]" } }, "tokenizers_version": "0.21.1", "unknown_token_counts": { "GPT-2": { "Bengali": 0, "English": 0, "Hindi": 0, "Marathi": 0, "Tamil": 0, "Telugu": 0 }, "Qwen2.5": { "Bengali": 0, "English": 0, "Hindi": 0, "Marathi": 0, "Tamil": 0, "Telugu": 0 }, "UMT5": { "Bengali": 0, "English": 0, "Hindi": 0, "Marathi": 0, "Tamil": 0, "Telugu": 0 }, "XLM-R": { "Bengali": 0, "English": 0, "Hindi": 0, "Marathi": 0, "Tamil": 0, "Telugu": 0 }, "mBERT": { "Bengali": 412, "English": 9, "Hindi": 86, "Marathi": 0, "Tamil": 4, "Telugu": 37 } } }